CN112069325A - Big data processing method based on block chain offline payment and cloud service pushing platform - Google Patents

Abstract

The embodiment of the application provides a big data processing method based on block chain offline payment and a cloud service push platform, mineable target services of service tags to be mined under target payment environment elements corresponding to an offline bill data set are considered, then mineable target services under all the target payment environment elements are grouped based on a preset subscribed push group, so that the difference between different target payment environment elements and the subscribed push group is considered, and therefore big data mining is performed on knowledge graph data sets of each subscribed push group based on push service graphs corresponding to the subscribed push group, the accuracy of big data mining can be effectively improved, and a big data mining result can be matched with an actual service scene better.

Description

Translated fromChinese

基于区块链离线支付的大数据处理方法及云服务推送平台Big data processing method and cloud service push platform based on blockchain offline payment

技术领域technical field

本申请涉及大数据技术领域，具体而言，涉及一种基于区块链离线支付的大数据处理方法及云服务推送平台。The present application relates to the field of big data technology, in particular, to a big data processing method and cloud service push platform based on blockchain offline payment.

背景技术Background technique

随着移动互联网技术和数字货币运营的发展，数字货币会逐渐成为作为今后新的主力支付方式，不仅可以支持在线支付，也可以如当前的现金交易一般支持离线网络状态下的离线支付。With the development of mobile Internet technology and digital currency operation, digital currency will gradually become the new main payment method in the future, which can not only support online payment, but also support offline payment in offline network state like current cash transactions.

由于在离线支付状态下，支付过程中产生的各种业务账单数据并不会实时同步到云服务平台中，而离线支付的场景同样也可以反映线下广泛用户的行为特征，因此仍旧需要对离线支付场景的离线账单数据集合进行大数据挖掘，以便于根据大数据级的分析结果改进后续的业务推送服务。Because in the offline payment state, the various business bill data generated during the payment process will not be synchronized to the cloud service platform in real time, and the offline payment scenario can also reflect the behavior characteristics of a wide range of offline users. The offline bill data collection of the payment scenario is used for big data mining, so as to improve the follow-up business push service according to the analysis results of the big data level.

然而，经本申请发明人研究发现，常规的大数据挖掘设计中，未考虑到实际的支付环境元素（例如支付业务场景类型、支付用户的用户类型等），导致大数据挖掘的精确性不高。However, the inventors of the present application found that, in the conventional big data mining design, the actual payment environment elements (such as payment business scenario types, payment user types, etc.) were not considered, resulting in low accuracy of big data mining. .

发明内容SUMMARY OF THE INVENTION

为了至少克服现有技术中的上述不足，本申请的目的在于提供一种基于区块链离线支付的大数据处理方法及云服务推送平台，通过考虑待挖掘服务标签在离线账单数据集合所对应的目标支付环境元素下的可挖掘目标服务，然后基于预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，从而考虑到不同目标支付环境元素和已订阅推送分组的差异，由此基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘，可以有效提高大数据挖掘的精确性，使得大数据挖掘结果更能够匹配实际的业务场景。In order to at least overcome the above-mentioned deficiencies in the prior art, the purpose of this application is to provide a big data processing method and cloud service push platform based on blockchain offline payment, by considering the service tag to be mined in the offline bill data set corresponding to Minable target services under the target payment environment element, and then group the mineable target services under each target payment environment element based on the predetermined subscribed push grouping, so as to take into account the differences between different target payment environment elements and subscribed push groups, Therefore, based on the push service portrait corresponding to the subscribed push group, big data mining is performed on the knowledge graph data set of each subscribed push group, which can effectively improve the accuracy of big data mining and make the big data mining results more suitable for actual business. Scenes.

第一方面，本申请提供一种基于区块链离线支付的大数据处理方法，应用于云服务推送平台，所述云服务推送平台与多个数字金融服务终端通信连接，所述方法包括：In the first aspect, the present application provides a big data processing method based on blockchain offline payment, which is applied to a cloud service push platform, wherein the cloud service push platform is connected to a plurality of digital financial service terminals in communication, and the method includes:

从每个数字金融服务终端中获取每个数字金融服务终端在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素；Obtain from each digital financial service terminal the offline bill data set generated by each digital financial service terminal in the blockchain offline payment environment and the target payment environment element corresponding to the offline bill data set;

获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合；Acquire the minable target services with the service tag to be mined under the target payment environment element, and group the mineable target services under each target payment environment element according to the predetermined subscribed push grouping, and generate each subscribed push grouping respectively A collection of minable target services;

针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。For each subscribed push group, acquire the knowledge graph data of each minable target service in the subscribed push group's minable target service set that matches the offline billing data set, and based on the push service corresponding to the subscribed push group The portrait performs big data mining on the knowledge graph data set of each subscribed push group.

在第一方面的一种可能的实现方式中，所述获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据的步骤，包括：In a possible implementation manner of the first aspect, the step of acquiring the knowledge graph data of each minable target service in the set of minable target services of the subscribed push group that matches the set of offline billing data includes: :

获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务相关的匹配关键词向量；Obtain the matching keyword vector related to each minable target service in the set of minable target services of the subscribed push group;

根据所述每个可挖掘目标服务相关的匹配关键词向量从所述离线账单数据集合中匹配对应的账单板块内容；Match the corresponding billing section content from the offline billing data set according to the matching keyword vector related to each mineable target service;

根据所述每个可挖掘目标服务相关的匹配关键词向量匹配的账单板块内容中每个业务记录板块对应的知识图谱内容，确定该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据。According to the knowledge graph content corresponding to each business record section in the billing section content matched by the matching keyword vector related to each minable target service, determine each minable target in the set of minable target services of the subscribed push group The service matches the knowledge graph data of the offline billing data set.

在第一方面的一种可能的实现方式中，所述方法还包括：In a possible implementation manner of the first aspect, the method further includes:

在大数据挖掘过程中判断是否存在用于表示可挖掘目标服务存在扩展加载业务的扩展加载业务信息，并在检测到所述扩展加载业务信息时，提取大数据挖掘的所述扩展加载业务信息对应的第一可挖掘目标服务的第一知识图谱以及与所述第一可挖掘目标服务存在扩展加载业务关系的至少一个第二可挖掘目标服务的第二知识图谱；In the process of big data mining, it is judged whether there is extended loading service information indicating that the target service can be mined and the extended loading service exists, and when the extended loading service information is detected, the corresponding information of the extended loading service information of big data mining is extracted. the first knowledge graph of the first mineable target service and the second knowledge graph of at least one second mineable target service that has an extended loading business relationship with the first mineable target service;

根据预设人工智能模型确定所述第一知识图谱和至少一个第二知识图谱之间的全局大数据挖掘信息。Global big data mining information between the first knowledge graph and at least one second knowledge graph is determined according to a preset artificial intelligence model.

在第一方面的一种可能的实现方式中，所述根据预设人工智能模型确定所述第一知识图谱和至少一个第二知识图谱之间的全局大数据挖掘信息的步骤，包括：In a possible implementation manner of the first aspect, the step of determining global big data mining information between the first knowledge graph and at least one second knowledge graph according to a preset artificial intelligence model includes:

将所述第一知识图谱按照每个相同的知识图谱节点与至少一个第二知识图谱对应的知识图谱节点进行融合后，得到融合知识图谱；After the first knowledge graph is fused with at least one knowledge graph node corresponding to the second knowledge graph according to each identical knowledge graph node, a fused knowledge graph is obtained;

将所述第一知识图谱和至少一个第二知识图谱添加到预设的数据地图分类队列，并基于所述数据地图分类队列建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数；The first knowledge map and at least one second knowledge map are added to a preset data map classification queue, and a plurality of first data map classification parameters and all the first data map classification parameters of the first knowledge map are established based on the data map classification queue. multiple second data map classification parameters of the second knowledge map;

根据每个第一数据地图分类参数确定所述第一可挖掘目标服务的第一知识表达信息，并根据每个第二数据地图分类参数确定所述第二可挖掘目标服务的第二知识表达信息，而后将所述第一知识表达信息和所述第二知识表达信息映射至知识实体特征模型，得到所述第一知识表达信息对应的第一知识图谱特征以及所述第二知识表达信息对应的第二知识图谱特征，并确定所述知识实体特征模型对应于所述融合知识图谱的多个知识语料对象，对所述多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表，针对每个知识语料挖掘列表，在预设的大数据挖掘进程中挖掘所述知识语料挖掘列表中的每个知识语料对象对应所述第一知识图谱特征的第一语料画像刻画特征和对应所述第二知识图谱特征的第二语料画像刻画特征；The first knowledge expression information of the first mineable target service is determined according to each first data map classification parameter, and the second knowledge expression information of the second mineable target service is determined according to each second data map classification parameter , and then map the first knowledge expression information and the second knowledge expression information to the knowledge entity feature model, and obtain the first knowledge map feature corresponding to the first knowledge expression information and the corresponding second knowledge expression information. the second knowledge graph feature, and determining that the knowledge entity feature model corresponds to a plurality of knowledge corpus objects of the fused knowledge graph, and summarizing the plurality of knowledge corpus objects to obtain at least a plurality of knowledge corpus mining lists of different categories, For each knowledge corpus mining list, in a preset big data mining process, mine each knowledge corpus object in the knowledge corpus mining list corresponding to the first corpus portrait feature of the first knowledge graph feature and the corresponding The second corpus portrait of the second knowledge graph feature depicts the feature;

根据所述知识语料挖掘列表中的每个知识语料对象对应的第一语料画像刻画特征和第二语料画像刻画特征的挖掘结果，按照知识预料的预设优先级进行拼接生成的模拟挖掘流，对所述拼接生成的模拟挖掘流进行还原，确定所述第一可挖掘目标服务和至少一个第二可挖掘目标服务的全局大数据挖掘信息。According to the mining results of the first corpus portrait description feature and the second corpus portrait description feature corresponding to each knowledge corpus object in the knowledge corpus mining list, the simulated mining flow generated by splicing is performed according to the preset priority of knowledge expectation, The simulated mining flow generated by the splicing is restored, and the global big data mining information of the first mineable target service and the at least one second mineable target service is determined.

在第一方面的一种可能的实现方式中，所述将所述第一知识图谱和至少一个第二知识图谱添加到预设的数据地图分类队列的步骤，包括：In a possible implementation manner of the first aspect, the step of adding the first knowledge graph and at least one second knowledge graph to a preset data map classification queue includes:

确定所述数据地图分类队列的扩展挖掘配置信息；其中，所述扩展挖掘配置信息用于表征所述数据地图分类队列对先后添加到的知识图谱进行处理时所分配的数据地图分类目标，所述数据地图分类目标用于表征所述数据地图分类队列对添加到的知识图谱进行挖掘时的挖掘特征节点信息；Determine the extended mining configuration information of the data map classification queue; wherein, the extended mining configuration information is used to represent the data map classification target allocated when the data map classification queue processes the knowledge maps added successively, the The data map classification target is used to represent the mining feature node information when the data map classification queue mines the added knowledge map;

基于所述扩展挖掘配置信息，确定将所述第一知识图谱添加到所述数据地图分类队列所对应的第一挖掘特征节点信息以及将所述第二知识图谱添加到所述数据地图分类队列所对应的第二挖掘特征节点信息；Based on the extended mining configuration information, it is determined that adding the first knowledge graph to the first mining feature node information corresponding to the data map classification queue and adding the second knowledge graph to the data map classification queue Corresponding second mining feature node information;

根据所述第一挖掘特征节点信息和所述第二挖掘特征节点信息确定在将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列时是否存在扩展加载业务；其中，所述扩展加载业务用于表征所述数据地图分类队列的挖掘存在扩展加载的响应行为；Determine, according to the first mining feature node information and the second mining feature node information, whether there is an extended loading service when the first knowledge graph and the second knowledge graph are added to the data map classification queue; wherein , the extended loading service is used to characterize that the mining of the data map classification queue has a response behavior of extended loading;

若否，则对所述第二挖掘特征节点信息进行调整得到第三挖掘特征节点信息，并基于所述第一挖掘特征节点信息和所述第三挖掘特征节点信息将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列，其中，所述第三挖掘特征节点信息与所述第二挖掘特征节点信息之间的特征差距与所述第一挖掘特征节点信息和所述第二挖掘特征节点信息之间的特征差距匹配；If not, the second mining feature node information is adjusted to obtain third mining feature node information, and the first knowledge graph and the third mining feature node information are combined based on the first mining feature node information and the third mining feature node information. The second knowledge graph is added to the data map classification queue, wherein the feature gap between the third mining feature node information and the second mining feature node information is the same as the first mining feature node information and all Describe the feature gap matching between the second mining feature node information;

若是，则持续采用所述第一挖掘特征节点信息和所述第二挖掘特征节点信息将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列。If so, the first knowledge graph and the second knowledge graph are added to the data map classification queue by continuously using the first mining feature node information and the second mining feature node information.

在第一方面的一种可能的实现方式中，基于所述数据地图分类队列建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数的步骤，包括：In a possible implementation manner of the first aspect, a plurality of first data map classification parameters of the first knowledge graph and a plurality of second data maps of the second knowledge graph are established based on the data map classification queue Steps to classify parameters, including:

基于所述数据地图分类队列确定所述第一知识图谱的第一挖掘节点序列以及所述第二知识图谱的第二挖掘节点序列；其中，所述挖掘节点序列用于表征知识图谱在不同挖掘节点下的挖掘业务关系；The first mining node sequence of the first knowledge graph and the second mining node sequence of the second knowledge graph are determined based on the data map classification queue; wherein the mining node sequence is used to represent the knowledge graph at different mining nodes Mining business relationships under the

分别根据所述第一挖掘节点序列以及所述第二挖掘节点序列在所述数据地图分类队列中建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数。Establish a plurality of first data map classification parameters of the first knowledge graph and multiple data map classification parameters of the second knowledge graph in the data map classification queue according to the first mining node sequence and the second mining node sequence, respectively. A second data map classification parameter.

在第一方面的一种可能的实现方式中，所述根据每个第一数据地图分类参数确定所述第一可挖掘目标服务的第一知识表达信息，并根据每个第二数据地图分类参数确定所述第二可挖掘目标服务的第二知识表达信息的步骤，包括：In a possible implementation manner of the first aspect, the first knowledge expression information of the first mineable target service is determined according to each first data map classification parameter, and the first knowledge expression information of the first mineable target service is determined according to each second data map classification parameter The step of determining the second knowledge representation information of the second mineable target service includes:

根据每个第一数据地图分类参数中的多个挖掘节点以及每相邻两个挖掘节点之间的挖掘画像地图参数确定每个第一数据地图分类参数对应的挖掘节点业务位图；Determine the mining node business bitmap corresponding to each first data map classification parameter according to the plurality of mining nodes in each first data map classification parameter and the mining portrait map parameters between every two adjacent mining nodes;

基于所述挖掘节点业务位图确定所述第一可挖掘目标服务的第一知识表达信息；其中，所述第一数据地图分类参数中的每个挖掘节点对应设置有挖掘画像地图索引参数，所述挖掘画像地图索引参数与任意一个挖掘节点的挖掘画像地图索引参数之间的匹配参数作为对应的挖掘画像地图参数，所述挖掘画像地图索引参数根据所述挖掘节点在所述第一数据地图分类参数中的挖掘频繁项模式确定；Determine the first knowledge expression information of the first mineable target service based on the mining node business bitmap; wherein, each mining node in the first data map classification parameter is correspondingly set with a mining portrait map index parameter, so The matching parameter between the mining portrait map index parameter and the mining portrait map index parameter of any mining node is used as the corresponding mining portrait map parameter, and the mining portrait map index parameter is classified according to the mining node in the first data map The mining frequent item pattern in the parameter is determined;

将每个第二数据地图分类参数的挖掘节点和挖掘节点对应的挖掘画像地图索引参数列出，得到每个第二数据地图分类参数对应的第一定位知识意图和第二定位知识意图；其中，所述第一定位知识意图为第二数据地图分类参数的挖掘节点对应的定位知识意图，所述第二定位知识意图为第二数据地图分类参数的挖掘画像地图索引参数对应的定位知识意图；The mining node of each second data map classification parameter and the mining portrait map index parameter corresponding to the mining node are listed, and the first positioning knowledge intention and the second positioning knowledge intention corresponding to each second data map classification parameter are obtained; wherein, The first positioning knowledge intention is the positioning knowledge intention corresponding to the mining node of the second data map classification parameter, and the second positioning knowledge intention is the positioning knowledge intention corresponding to the mining portrait map index parameter of the second data map classification parameter;

确定所述第一定位知识意图相对于所述第二定位知识意图的第一意图抽取实体以及所述第二定位知识意图相对于所述第二定位知识意图的第二意图抽取实体；determining a first intent extraction entity of the first location knowledge intent relative to the second location knowledge intent and a second intent extraction entity of the second location knowledge intent relative to the second location knowledge intent;

获取所述第一意图抽取实体和所述第二意图抽取实体中具有相同的抽取实体连续性的至少三个目标抽取实体节点，并根据所述目标抽取实体节点确定出所述第二数据地图分类参数的第二知识表达信息；其中，所述抽取实体连续性用于表征每两个抽取实体之间的实体循环关系。Obtain at least three target extraction entity nodes with the same extraction entity continuity in the first intent extraction entity and the second intent extraction entity, and determine the second data map classification according to the target extraction entity nodes The second knowledge expression information of the parameter; wherein, the extracted entity continuity is used to represent the entity cycle relationship between every two extracted entities.

在第一方面的一种可能的实现方式中，所述对所述多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表的步骤，包括：In a possible implementation manner of the first aspect, the step of aggregating the multiple knowledge corpus objects to obtain at least multiple knowledge corpus mining lists of different categories includes:

确定所述知识实体特征模型中的每个知识语料对象对应的知识图谱特征的业务标签范围；Determine the business label range of the knowledge graph feature corresponding to each knowledge corpus object in the knowledge entity feature model;

确定每个知识语料对象对应的知识图谱特征的图谱节点重合范围；其中，所述图谱节点重合范围为每个知识语料对象对应的知识图谱特征中第一知识图谱特征与第二知识图谱特征的重合部分；Determine the overlap range of the graph nodes of the knowledge graph feature corresponding to each knowledge corpus object; wherein, the overlap range of the graph nodes is the overlap of the first knowledge graph feature and the second knowledge graph feature in the knowledge graph feature corresponding to each knowledge corpus object part;

确定每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的融合挖掘信息；其中，所述融合挖掘信息通过对第一知识图谱特征和第二知识图谱特征对应设定业务标签范围的挖掘指向对象进行特征参数并集计算得到；Determine the fusion mining information of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object; wherein, the fusion mining information is set by correspondingly setting the business label range for the first knowledge graph feature and the second knowledge graph feature The mining points to the object to calculate the feature parameter union;

根据每个知识语料对象对应的知识图谱特征的业务标签范围、图谱节点重合范围和融合挖掘信息确定每个知识语料对象的结构化主题特征序列；Determine the structured topic feature sequence of each knowledge corpus object according to the business label range, graph node coincidence range and fusion mining information of the knowledge graph feature corresponding to each knowledge corpus object;

基于每个知识语料对象的结构化主题特征序列对每个知识语料对象进行汇总，得到所述至少多个不同类别的知识语料挖掘列表。Summarize each knowledge corpus object based on the structured topic feature sequence of each knowledge corpus object to obtain the at least multiple knowledge corpus mining lists of different categories.

在第一方面的一种可能的实现方式中，所述在预设的大数据挖掘进程中挖掘所述知识语料挖掘列表中的每个知识语料对象对应所述第一知识图谱特征的第一语料画像刻画特征和对应所述第二知识图谱特征的第二语料画像刻画特征的步骤，包括：In a possible implementation manner of the first aspect, the mining of each knowledge corpus object in the knowledge corpus mining list in a preset big data mining process corresponds to the first corpus of the first knowledge graph feature The step of portraying the features of the portrait and the second corpus portrait corresponding to the second knowledge map feature, including:

确定每个知识语料挖掘列表中每个知识语料对象对应的结构化主题特征序列的扩展挖掘配置信息；Determine the extended mining configuration information of the structured topic feature sequence corresponding to each knowledge corpus object in each knowledge corpus mining list;

根据所述扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差；其中，所述数据地图分类误差用于表征每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的挖掘误差情况；The data map classification error of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object in each summary is determined according to the extended mining configuration information; wherein, the data map classification error is used to represent each Mining errors of the first knowledge graph feature and the second knowledge graph feature corresponding to the knowledge corpus object;

判断每个数据地图分类误差与所述大数据挖掘进程对应的基准挖掘误差的差值是否在预设差值区间内；其中，所述预设差值区间用于表征大数据挖掘进程处于正常运行时每个数据地图分类误差所处的区间；Determine whether the difference between the classification error of each data map and the benchmark mining error corresponding to the big data mining process is within a preset difference interval; wherein the preset difference interval is used to indicate that the big data mining process is in normal operation is the interval in which the classification error of each data map is located;

在每个数据地图分类误差与所述大数据挖掘进程对应的基准同步系数的差值均落入所述预设差值区间时，基于所述大数据挖掘进程模拟验证所述知识语料挖掘列表中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征；When the difference between each data map classification error and the reference synchronization coefficient corresponding to the big data mining process falls within the preset difference interval, simulate and verify that the knowledge corpus mining list is based on the big data mining process The first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object of ;

否则，根据所述大数据挖掘进程的参数更新子进程对未落入所述预设差值区间内的差值对应的数据地图分类误差对应的扩展挖掘配置信息进行修正，并返回根据所述扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差的步骤。Otherwise, correct the extended mining configuration information corresponding to the data map classification error corresponding to the difference value that does not fall within the preset difference value interval according to the parameter update sub-process of the big data mining process, and return the data according to the extended mining configuration information. The step of mining the configuration information to determine the data map classification error of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object in each summary.

第二方面，本申请实施例还提供一种基于区块链离线支付的大数据处理装置，应用于云服务推送平台，所述云服务推送平台与多个数字金融服务终端通信连接，所述装置包括：In the second aspect, the embodiments of the present application further provide a big data processing device based on blockchain offline payment, which is applied to a cloud service push platform, where the cloud service push platform is connected to a plurality of digital financial service terminals in communication, and the device include:

获取模块，用于从每个数字金融服务终端中获取每个数字金融服务终端在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素；an acquisition module, configured to acquire, from each digital financial service terminal, an offline bill data set generated by each digital financial service terminal under the blockchain offline payment environment and a target payment environment element corresponding to the offline bill data set;

分组模块，用于获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合；The grouping module is used to obtain the minable target services of the service tag to be mined under the target payment environment element, and group the minable target services under each target payment environment element according to the predetermined subscribed push grouping, and generate each A set of minable target services that have subscribed to push groups;

大数据挖掘模块，用于针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。The big data mining module is used to obtain, for each subscribed push group, the knowledge graph data of each minable target service in the subscribed push group's minable target service set that matches the offline billing data set, and based on the The push service portrait corresponding to the subscription push group performs big data mining on the knowledge graph data set of each subscribed push group.

第三方面，本申请实施例还提供一种基于区块链离线支付的大数据处理系统，所述基于区块链离线支付的大数据处理系统包括云服务推送平台以及与所述云服务推送平台通信连接的多个数字金融服务终端；In a third aspect, the embodiments of the present application further provide a big data processing system based on blockchain offline payment. The big data processing system based on blockchain offline payment includes a cloud service push platform and a cloud service push platform. Multiple digital financial service terminals connected by communication;

所述云服务推送平台，用于从每个数字金融服务终端中获取每个数字金融服务终端在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素；The cloud service push platform is used to obtain, from each digital financial service terminal, the offline bill data set generated by each digital financial service terminal in the blockchain offline payment environment and the target payment corresponding to the offline bill data set environmental elements;

所述云服务推送平台，用于获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合；The cloud service push platform is used to obtain the minable target services under the target payment environment element with the service tag to be mined, and group the minable target services under each target payment environment element according to a predetermined subscribed push grouping , respectively generate a set of minable target services for each subscribed push group;

所述云服务推送平台，用于针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。The cloud service push platform is configured to, for each subscribed push group, acquire knowledge graph data of each minable target service in the subscribed push group's minable target service set that matches the offline billing data set, and Based on the push service portrait corresponding to the subscribed push group, big data mining is performed on the knowledge graph data set of each subscribed push group.

第四方面，本申请实施例还提供一种云服务推送平台，所述云服务推送平台包括处理器、机器可读存储介质和网络接口，所述机器可读存储介质、所述网络接口以及所述处理器之间通过总线系统相连，所述网络接口用于与至少一个数字金融服务终端通信连接，所述机器可读存储介质用于存储程序、指令或代码，所述处理器用于执行所述机器可读存储介质中的程序、指令或代码，以执行第一方面或者第一方面中任意一个可能的实现方式中的基于区块链离线支付的大数据处理方法。In a fourth aspect, embodiments of the present application further provide a cloud service push platform, the cloud service push platform includes a processor, a machine-readable storage medium, and a network interface, the machine-readable storage medium, the network interface, and all The processors are connected through a bus system, the network interface is used to communicate with at least one digital financial service terminal, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the A program, instruction or code in a machine-readable storage medium to execute the big data processing method based on blockchain offline payment in the first aspect or any possible implementation manner of the first aspect.

第五方面，本申请实施例提供一种计算机可读存储介质，计算机可读存储介质中存储有指令，当其被执行时，使得计算机执行上述第一方面或者第一方面中任意一个可能的实现方式中的基于区块链离线支付的大数据处理方法。In a fifth aspect, embodiments of the present application provide a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when executed, cause a computer to execute the first aspect or any possible implementation of the first aspect The big data processing method based on blockchain offline payment in the method.

基于上述任意一个方面，本申请通过考虑待挖掘服务标签在离线账单数据集合所对应的目标支付环境元素下的可挖掘目标服务，然后基于预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，从而考虑到不同目标支付环境元素和已订阅推送分组的差异，由此基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘，可以有效提高大数据挖掘的精确性，使得大数据挖掘结果更能够匹配实际的业务场景。Based on any of the above aspects, the present application considers the minable target services of the service tag to be mined under the target payment environment element corresponding to the offline billing data set, and then, based on the predetermined subscribed push grouping, analyzes the available target services under each target payment environment element. Mining target services for grouping, so as to take into account the differences between different target payment environment elements and subscribed push groups, based on the push service portraits corresponding to the subscribed push groups, perform big data mining on the knowledge graph data set of each subscribed push group , which can effectively improve the accuracy of big data mining, so that the results of big data mining can better match the actual business scenarios.

附图说明Description of drawings

为了更清楚地说明本申请实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本申请的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其它相关的附图。In order to illustrate the technical solutions of the embodiments of the present application more clearly, the following drawings will briefly introduce the drawings that need to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore do not It should be regarded as a limitation of the scope. For those skilled in the art, other related drawings can also be obtained according to these drawings without any creative effort.

图1为本申请实施例提供的基于区块链离线支付的大数据处理系统的应用场景示意图；1 is a schematic diagram of an application scenario of a big data processing system based on blockchain offline payment provided by an embodiment of the present application;

图2为本申请实施例提供的基于区块链离线支付的大数据处理方法的流程示意图；2 is a schematic flowchart of a big data processing method based on blockchain offline payment provided by an embodiment of the present application;

图3为本申请实施例提供的基于区块链离线支付的大数据处理装置的功能模块示意图；3 is a schematic diagram of functional modules of a big data processing device based on blockchain offline payment provided by an embodiment of the present application;

图4为本申请实施例提供的用于实现上述的基于区块链离线支付的大数据处理方法的云服务推送平台的结构组件示意框图。FIG. 4 is a schematic block diagram of the structural components of the cloud service push platform for implementing the above-mentioned blockchain-based offline payment big data processing method according to an embodiment of the present application.

具体实施方式Detailed ways

下面结合说明书附图对本申请进行具体说明，方法实施例中的具体操作方法也可以应用于装置实施例或系统实施例中。The present application will be specifically described below with reference to the accompanying drawings in the specification, and the specific operation methods in the method embodiments may also be applied to the apparatus embodiments or the system embodiments.

图1是本申请一种实施例提供的基于区块链离线支付的大数据处理系统10的交互示意图。基于区块链离线支付的大数据处理系统10可以包括云服务推送平台100以及与云服务推送平台100通信连接的数字金融服务终端200。图1所示的基于区块链离线支付的大数据处理系统10仅为一种可行的示例，在其它可行的实施例中，该基于区块链离线支付的大数据处理系统10也可以仅包括图1所示组成部分的其中一部分或者还可以包括其它的组成部分。FIG. 1 is an interactive schematic diagram of a bigdata processing system 10 based on blockchain offline payment provided by an embodiment of the present application. The bigdata processing system 10 based on blockchain offline payment may include a cloudservice push platform 100 and a digitalfinancial service terminal 200 connected in communication with the cloudservice push platform 100 . The bigdata processing system 10 based on blockchain offline payment shown in FIG. 1 is only a feasible example. In other feasible embodiments, the bigdata processing system 10 based on blockchain offline payment may also only include Some of the components shown in FIG. 1 may also include other components.

本实施例中，数字金融服务终端200可以包括移动设备、平板计算机、膝上型计算机等或其任意组合。在一些实施例中，移动设备可以包括物联网设备、可穿戴设备、智能移动设备、虚拟现实设备、或增强现实设备等，或其任意组合。在一些实施例中，物联网设备可以包括智能电器设备的控制设备、智能监控设备、智能电视、智能摄像机等，或其任意组合。在一些实施例中，可穿戴设备可包括智能手环、智能鞋带、智能玻璃、智能头盔、智能手表、智能服装、智能背包、智能配件等，或其任何组合。在一些实施例中，智能移动设备可以包括智能手机、个人数字助理、游戏设备等，或其任意组合。在一些实施例中，虚拟现实设备和增强现实设备可以包括虚拟现实头盔、虚拟现实玻璃、虚拟现实贴片、增强现实头盔、增强现实玻璃、或增强现实贴片等，或其任意组合。例如，虚拟现实设备和增强现实设备可以包括各种虚拟现实产品等。In this embodiment, the digitalfinancial service terminal 200 may include a mobile device, a tablet computer, a laptop computer, etc., or any combination thereof. In some embodiments, the mobile device may include an Internet of Things device, a wearable device, a smart mobile device, a virtual reality device, or an augmented reality device, etc., or any combination thereof. In some embodiments, the IoT devices may include control devices of smart electrical devices, smart monitoring devices, smart TVs, smart cameras, etc., or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, smart shoelaces, smart glass, smart helmets, smart watches, smart clothing, smart backpacks, smart accessories, etc., or any combination thereof. In some embodiments, a smart mobile device may include a smartphone, a personal digital assistant, a gaming device, etc., or any combination thereof. In some embodiments, virtual reality devices and augmented reality devices may include virtual reality helmets, virtual reality glasses, virtual reality patches, augmented reality helmets, augmented reality glasses, or augmented reality patches, or the like, or any combination thereof. For example, virtual reality devices and augmented reality devices may include various virtual reality products and the like.

本实施例中，基于区块链离线支付的大数据处理系统10中的云服务推送平台100和数字金融服务终端200可以通过配合执行以下方法实施例所描述的基于区块链离线支付的大数据处理方法，具体云服务推送平台100和数字金融服务终端200的执行步骤部分可以参照以下方法实施例的详细描述。In this embodiment, the cloudservice push platform 100 and the digitalfinancial service terminal 200 in the bigdata processing system 10 based on blockchain offline payment can cooperate to execute the big data based on blockchain offline payment described in the following method embodiments. For the processing method, for the specific execution steps of the cloudservice push platform 100 and the digitalfinancial service terminal 200, reference may be made to the detailed description of the following method embodiments.

基于本申请提供的技术方案的发明构思出发，本申请提供的云服务推送平台100可以应用在例如智慧医疗、智慧城市管理、智慧工业互联网、通用业务监控管理等可以应用大数据技术或者是云计算技术等的场景中，再比如，还可以应用在包括但不限于新能源汽车系统管理、智能云办公、云平台数据处理、云游戏数据处理、云直播处理、云汽车管理平台、区块链金融数据服务平台等，但不限于此。Based on the inventive concept of the technical solution provided in this application, the cloudservice push platform 100 provided in this application can be applied to, for example, smart medical care, smart city management, smart industrial Internet, general business monitoring and management, etc., where big data technology or cloud computing can be applied. In scenarios such as technology, for example, it can also be applied to include but not limited to new energy vehicle system management, intelligent cloud office, cloud platform data processing, cloud game data processing, cloud live broadcast processing, cloud car management platform, blockchain finance Data service platform, etc., but not limited to this.

为了解决前述背景技术中的技术问题，图2为本申请实施例提供的基于区块链离线支付的大数据处理方法的流程示意图，本实施例提供的基于区块链离线支付的大数据处理方法可以由图1中所示的云服务推送平台100执行，下面对该基于区块链离线支付的大数据处理方法进行详细介绍。In order to solve the technical problems in the aforementioned background art, FIG. 2 is a schematic flowchart of the big data processing method based on blockchain offline payment provided by the embodiment of the application, and the big data processing method based on blockchain offline payment provided by this embodiment. It can be executed by the cloudservice push platform 100 shown in FIG. 1 , and the big data processing method based on blockchain offline payment will be described in detail below.

步骤S110，从每个数字金融服务终端200中获取每个数字金融服务终端200在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素。In step S110, the offline bill data set generated by each digitalfinancial service terminal 200 in the blockchain offline payment environment and the target payment environment element corresponding to the offline bill data set are obtained from each digitalfinancial service terminal 200.

其中，目标支付环境元素可以用于表示具体支付过程中离线获取的环境元素，例如支付业务场景类型、支付用户的用户类型等。The target payment environment element may be used to represent an environment element obtained offline in a specific payment process, such as the payment service scenario type, the user type of the payment user, and the like.

步骤S120，获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合。Step S120: Acquire the minable target services of the service tag to be mined under the target payment environment element, and group the minable target services under each target payment environment element according to the predetermined subscribed push grouping, and generate each of the minable target services respectively. A collection of minable target services to subscribe to push groups.

其中，离线账单数据可以是指对于每一次离线支付场景而言，通常会包括多个账单统计板块，例如可以包括但不限于账单内容、账单场景、账单收支情况等等。Wherein, the offline bill data may refer to each offline payment scenario, which usually includes multiple bill statistics sections, such as but not limited to bill content, bill scene, bill income and expenditure, and the like.

其中，可挖掘目标服务可以用于表示待挖掘服务标签在每个离线账单数据的目标支付环境元素下的具体应用的挖掘类型标签，例如生鲜类知识新挖掘类型标签、数码产品类知识新挖掘类型标签等。Among them, the minable target service can be used to represent the mining type label of the specific application of the service label to be mined under the target payment environment element of each offline bill data, such as the new mining type label for fresh knowledge, and the new mining type for digital product knowledge. Type tags, etc.

本实施例中，预定的已订阅推送分组可以根据实际设计需求进行灵活选择，主要用于表征为不同用户提供的订阅推送选定菜单，在此不作详细限定。In this embodiment, the predetermined subscribed push group can be flexibly selected according to actual design requirements, and is mainly used to represent the subscription push selection menu provided for different users, which is not limited in detail here.

步骤S130，针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。Step S130, for each subscribed push group, obtain the knowledge graph data of each minable target service in the subscribed push group's minable target service set that matches the offline billing data set, and based on the subscribed push group corresponding The push service profile of the big data mining is carried out on the knowledge graph data collection of each subscribed push group.

基于上述步骤，本实施例通过考虑待挖掘服务标签在离线账单数据集合所对应的目标支付环境元素下的可挖掘目标服务，然后基于预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，从而考虑到不同目标支付环境元素和已订阅推送分组的差异，由此基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘，可以有效提高大数据挖掘的精确性，使得大数据挖掘结果更能够匹配实际的业务场景。Based on the above steps, this embodiment considers the minable target services under the target payment environment element corresponding to the offline billing data set of the service tag to be mined, and then, based on the predetermined subscribed push grouping, analyzes the mineable target service under each target payment environment element. The target services are grouped to take into account the differences between different target payment environment elements and the subscribed push groups. Based on the push service portraits corresponding to the subscribed push groups, big data mining is performed on the knowledge graph data set of each subscribed push group. It can effectively improve the accuracy of big data mining, so that the results of big data mining can better match the actual business scenarios.

在一种可能的实现方式中，譬如，针对步骤S120，为了提高划分的精确度，并且减少冗余信息以提高分组准确度，本实施例可以获取每个预定的已订阅推送分组所对应的订阅扩展特征样本，形成每个预定的已订阅推送分组的订阅扩展特征样本序列，并获取各个目标支付环境元素的每个目标订阅扩展特征样本与订阅扩展特征样本序列的订阅扩展特征样本的关联订阅扩展特征样本信息。In a possible implementation manner, for example, for step S120, in order to improve the accuracy of division and reduce redundant information to improve the accuracy of grouping, this embodiment may acquire the subscription corresponding to each predetermined subscribed push group Extend the feature samples, form the subscription extended feature sample sequence of each predetermined subscribed push group, and obtain the associated subscription extension of each target subscription extended feature sample of each target payment environment element and the subscription extended feature sample of the subscription extended feature sample sequence Feature sample information.

在此基础上，可以根据目标订阅扩展特征样本与订阅扩展特征样本序列的订阅扩展特征样本的关联订阅扩展特征样本信息，计算每种目标已订阅推送分组的关键订阅扩展特征样本的样本业务覆盖范围，并根据每种目标已订阅推送分组的关键订阅扩展特征样本的样本业务覆盖范围，从订阅扩展特征样本序列中选取订阅扩展特征样本，得到初始订阅扩展特征样本分布空间。On this basis, the sample business coverage of the key subscription extended feature samples of each target subscribed push group can be calculated according to the target subscription extended feature sample and the subscription extended feature sample information of the subscription extended feature sample sequence of the subscription extended feature sample sequence. , and according to the sample business coverage of the key subscription extended feature samples of each target subscribed push group, select the subscription extended feature samples from the subscription extended feature sample sequence to obtain the initial subscription extended feature sample distribution space.

在一种可能的示例中，若初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围大于总素材分布样本业务覆盖范围要求的最大总素材分布样本业务覆盖范围，则将初始订阅扩展特征样本分布空间中的第一关键订阅扩展特征样本分散到第一分布样本业务覆盖范围，并且将初始订阅扩展特征样本分布空间中的第二关键订阅扩展特征样本聚集到第一分布样本业务覆盖范围。In a possible example, if the total material distribution sample business coverage of the initial subscription extended feature sample distribution space is greater than the maximum total material distribution sample business coverage required by the total material distribution sample business coverage, the initial subscription extended feature sample service coverage The first key subscription extended feature samples in the distribution space are scattered to the service coverage of the first distribution sample, and the second key subscription extended feature samples in the initial subscription extended feature sample distribution space are aggregated to the first distribution sample service coverage.

其中，值得说明的是，第二关键订阅扩展特征样本可以是指关键订阅扩展特征样本在所在的订阅地图的单位密集程度小于设定程度的关键订阅扩展特征样本，第一关键订阅扩展特征样本可以是指关键订阅扩展特征样本在所在的订阅地图的单位密集程度不小于设定程度的关键订阅扩展特征样本，第一分布样本业务覆盖范围可根据实际需求进行设定，但是第一分部样本业务覆盖范围不应当与总素材分布样本业务覆盖范围要求的最大总素材分布样本业务覆盖范围相差过大。It is worth noting that the second key subscription extended feature sample may refer to a key subscription extended feature sample where the unit density of the subscription map where the key subscription extended feature sample is located is less than the set degree, and the first key subscription extended feature sample may be Refers to the key subscription extended feature samples in which the unit density of the subscription map is not less than the set level. The coverage of the first distribution sample business can be set according to actual needs, but the first division sample business The coverage should not be too different from the maximum total material distribution sample business coverage required by the total material distribution sample business coverage.

然后，计算本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围，若本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围大于最大总素材分布样本业务覆盖范围，则再一次对本次调整后的初始订阅扩展特征样本分布空间执行以上处理。Then, calculate the total material distribution sample business coverage of the initial subscription extended feature sample distribution space after this adjustment. If the total material distribution sample business coverage of the initial subscription extended feature sample distribution space after this adjustment is greater than the maximum total material distribution If the sample business coverage is exceeded, the above processing is performed again on the initial subscription extension feature sample distribution space after this adjustment.

再例如，若本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围小于或者等于最大总素材分布样本业务覆盖范围，则可以将本次调整前的初始订阅扩展特征样本分布空间作为第一调整分布空间，按照已订阅推送分组由低优先级到高优先级的顺序将各目标已订阅推送分组进行排序，得到目标已订阅推送分组序列。For another example, if the total material distribution sample business coverage of the initial subscription extension feature sample distribution space after this adjustment is less than or equal to the maximum total material distribution sample business coverage, the initial subscription extension feature sample distribution before this adjustment can be used. The space is used as the first adjustment distribution space, and the target subscribed push groups are sorted according to the order of the subscribed push groups from low priority to high priority, so as to obtain the target subscribed push group sequence.

在此基础上，可以根据目标已订阅推送分组序列对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合。On this basis, the minable target services under each target payment environment element can be grouped according to the target subscribed push grouping sequence, and a set of minable target services for each subscribed push grouping can be generated respectively.

譬如详细地，可以根据目标已订阅推送分组序列，将各目标已订阅推送分组进行聚类，每个聚类中包括与目标已订阅推送分组序列的订阅地图区域相关的、且与订阅地图区域的范围差异一致的第一已订阅推送分组和第二已订阅推送分组，第一已订阅推送分组的优先级小于第二已订阅推送分组。For example, in detail, each target subscribed push group can be clustered according to the target subscribed push group sequence. For the first subscribed push group and the second subscribed push group with the same range difference, the priority of the first subscribed push group is lower than that of the second subscribed push group.

然后，按照与订阅地图区域的范围差异由低优先级到高优先级的顺序，依次将每个聚类作为目标聚类，对目标聚类进行以下第二调整处理： 将第一调整分布空间中目标聚类的第一已订阅推送分组的关键订阅扩展特征样本增加设定数目，并且将第一调整分布空间中目标聚类的第二已订阅推送分组的关键订阅扩展特征样本减少设定数目。Then, according to the order of the range difference from the subscription map area from low priority to high priority, each cluster is taken as the target cluster in turn, and the following second adjustment processing is performed on the target cluster: The key subscription extended feature samples of the first subscribed push group of the target cluster are increased by a set number, and the key subscription extended feature samples of the second subscribed push group of the target cluster in the first adjustment distribution space are decreased by a set number.

在此基础上，可以判断本次调整后的第一调整分布空间的总素材分布样本业务覆盖范围是否大于总素材分布样本业务覆盖范围要求，若本次调整后的第一调整分布空间的总素材分布样本业务覆盖范围大于总素材分布样本业务覆盖范围要求，则将本次调整后的第一调整分布空间作为最终订阅扩展特征样本分布空间。若本次调整后的第一调整分布空间的总素材分布样本业务覆盖范围不大于总素材分布样本业务覆盖范围要求，则将下一个聚类作为新的目标聚类，对新的目标聚类进行第二调整处理。On this basis, it can be determined whether the total material distribution sample business coverage of the first adjustment distribution space after this adjustment is greater than the total material distribution sample business coverage requirement, if the total material distribution of the first adjustment distribution space after this adjustment is the total material distribution If the coverage of the distribution sample business is greater than the requirement of the total material distribution sample business coverage, the first adjusted distribution space after this adjustment is used as the final subscription extension feature sample distribution space. If the total material distribution sample business coverage of the first adjusted distribution space after this adjustment is not greater than the total material distribution sample business coverage requirement, the next cluster will be used as a new target cluster, and the new target cluster will be processed. The second adjustment process.

又例如，若初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围小于大于总素材分布样本业务覆盖范围要求的最小总素材分布样本业务覆盖范围，则对初始订阅扩展特征样本分布空间进行以下第三调整处理： 将初始订阅扩展特征样本分布空间中的第一关键订阅扩展特征样本增加第一分布样本业务覆盖范围，并且将初始订阅扩展特征样本分布空间中的第二关键订阅扩展特征样本减少第一分布样本业务覆盖范围。For another example, if the total material distribution sample business coverage of the initial subscription extended feature sample distribution space is smaller than the minimum total material distribution sample business coverage required by the total material distribution sample business coverage, then the initial subscription extended feature sample distribution space is performed as follows: The third adjustment process: increase the first key subscription extended feature sample in the distribution space of the initial subscription extended feature sample to the first distribution sample business coverage, and decrease the second key subscription extended feature sample in the initial subscription extended feature sample distribution space The first distribution sample business coverage.

在此基础上，计算本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围，若本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围小于最小总素材分布样本业务覆盖范围，则再一次对本次调整后的初始订阅扩展特征样本分布空间执行第三调整处理。或者，若本次调整后的初始订阅扩展特征样本分布空间的总素材分布样本业务覆盖范围大于或者等于最小总素材分布样本业务覆盖范围，则将本次调整前的初始订阅扩展特征样本分布空间作为第二调整分布空间，按照已订阅推送分组由低优先级到高优先级的顺序将各目标已订阅推送分组进行排序，得到目标已订阅推送分组序列。On this basis, calculate the total material distribution sample business coverage of the initial subscription extended feature sample distribution space after this adjustment, if the adjusted initial subscription extended feature sample distribution space The total material distribution sample business coverage is less than the minimum If the business coverage of the total material distribution samples is reached, the third adjustment process is performed again on the initially subscribed extended feature sample distribution space after this adjustment. Or, if the total material distribution sample business coverage of the initial subscription extended feature sample distribution space after this adjustment is greater than or equal to the minimum total material distribution sample business coverage, the initial subscription extended feature sample distribution space before this adjustment is used as The second adjusts the distribution space, and sorts the subscribed push groups of each target according to the order of the subscribed push groups from low priority to high priority, so as to obtain a sequence of target subscribed push groups.

由此，可以根据目标已订阅推送分组序列，将各目标已订阅推送分组进行聚类，每个聚类中包括在目标已订阅推送分组序列的订阅地图区域关联的、且与订阅地图区域的范围差异一致的第一已订阅推送分组和第二已订阅推送分组，第一已订阅推送分组的优先级小于第二已订阅推送分组。Thus, each target subscribed push group can be clustered according to the target subscribed push group sequence, and each cluster includes the range associated with the subscription map area of the target subscribed push group sequence and with the subscription map area For the first subscribed push group and the second subscribed push group with consistent differences, the priority of the first subscribed push group is lower than that of the second subscribed push group.

然后，按照与订阅地图区域的范围差异由低优先级到高优先级的顺序，依次将每个聚类作为目标聚类，对目标聚类进行以下第四调整处理：将第二调整分布空间中目标聚类的第一已订阅推送分组的关键订阅扩展特征样本减少设定数目，并且将第二调整分布空间中目标聚类的第二已订阅推送分组的关键订阅扩展特征样本增加设定数目。Then, according to the order of the range difference from the subscription map area from low priority to high priority, each cluster is taken as the target cluster in turn, and the following fourth adjustment processing is performed on the target cluster: The key subscription extended feature samples of the first subscribed push group of the target cluster are decreased by a set number, and the key subscription extended feature samples of the second subscribed push group of the target cluster in the second adjustment distribution space are increased by a set number.

进一步地，本实施例可以判断本次调整后的第二调整分布空间的总素材分布样本业务覆盖范围是否大于总素材分布样本业务覆盖范围要求，若本次调整后的第二调整分布空间的总素材分布样本业务覆盖范围大于总素材分布样本业务覆盖范围要求，则将本次调整后的第二调整分布空间作为最终订阅扩展特征样本分布空间，若本次调整后的第二调整分布空间的总素材分布样本业务覆盖范围不大于总素材分布样本业务覆盖范围要求，则将下一个聚类作为新的目标聚类，对新的目标聚类进行第四调整处理。Further, this embodiment can determine whether the total material distribution sample service coverage of the second adjustment distribution space after this adjustment is greater than the total material distribution sample service coverage requirement, if the total material distribution sample service coverage of the second adjustment distribution space after this adjustment is The material distribution sample business coverage is greater than the total material distribution sample business coverage requirement, then the second adjustment distribution space after this adjustment is used as the final subscription extension feature sample distribution space. If the total amount of the second adjustment distribution space after this adjustment is If the business coverage of the material distribution sample is not greater than the business coverage requirement of the total material distribution sample, the next cluster is taken as the new target cluster, and the fourth adjustment processing is performed on the new target cluster.

由此，可以将各个目标已订阅推送分组的最终订阅扩展特征样本分布空间中的每个订阅扩展特征样本的可挖掘目标服务分别归类为该已订阅推送分组的可挖掘目标服务集合。Therefore, each target service subscribed to the extended feature sample in the distribution space of the final subscription extended feature sample distribution space of each target subscribed push group can be classified as a set of minable target services of the subscribed push group.

在一种可能的实现方式中，针对步骤S130，在获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据的过程中，还可以通过以下示例性的子步骤来实现，详细描述如下。In a possible implementation manner, for step S130, in the process of acquiring the knowledge graph data of each minable target service in the set of minable target services of the subscribed push group that matches the set of offline billing data, further This can be achieved through the following exemplary sub-steps, which are described in detail below.

子步骤S131，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务相关的匹配关键词向量。Sub-step S131: Obtain a matching keyword vector related to each minable target service in the set of minable target services of the subscribed push group.

子步骤S132，根据每个可挖掘目标服务相关的匹配关键词向量从离线账单数据集合中匹配对应的账单板块内容。Sub-step S132: Match the corresponding billing section content from the offline billing data set according to the matching keyword vector related to each minable target service.

子步骤S133，根据每个可挖掘目标服务相关的匹配关键词向量匹配的账单板块内容中每个业务记录板块对应的知识图谱内容，确定该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于离线账单数据集合的知识图谱数据。Sub-step S133, according to the knowledge map content corresponding to each business record section in the billing section content matched by the matching keyword vector related to each minable target service, determine each minable target service set in the subscribed push grouping that can be mined. Mining the knowledge graph data of the target service matching the offline billing data collection.

在一种可能的实现方式中，譬如，针对步骤S130，在基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘的过程中，可以通过以下子步骤来实现。In a possible implementation manner, for example, for step S130, in the process of performing big data mining on the knowledge graph data set of each subscribed push group based on the push service portrait corresponding to the subscribed push group, the following steps may be used to perform big data mining. steps to achieve.

子步骤S134，基于已订阅推送分组对应的推送服务画像确定每个已订阅推送分组的每个推送服务画像节点的推送服务画像节点参数以及推送服务画像节点所覆盖的画像激活内容。Sub-step S134, based on the push service portrait corresponding to the subscribed push group, determine the push service portrait node parameters of each push service portrait node of each subscribed push group and the portrait activation content covered by the push service portrait node.

子步骤S135，根据每个已订阅推送分组中推送服务画像节点的推送服务画像节点参数以及推送服务画像节点所覆盖的画像激活内容确定每个已订阅推送分组中对推送服务画像节点进行大数据挖掘所需要的大数据挖掘组件的挖掘流程参数。Sub-step S135, according to the push service portrait node parameters of the push service portrait node in each subscribed push group and the portrait activation content covered by the push service portrait node, it is determined to perform big data mining on the push service portrait node in each subscribed push group The mining process parameters of the required big data mining components.

子步骤S136，根据每个推送服务画像节点所需要的大数据挖掘组件的挖掘流程参数，将每个大数据挖掘组件确定为一挖掘单位，该挖掘单位所对应的运行配置信息为该推送服务画像节点包含的当前已配置的已订阅推送分组的运行配置信息之外的运行配置信息。Sub-step S136, according to the mining process parameters of the big data mining components required by each push service portrait node, determine each big data mining component as a mining unit, and the operation configuration information corresponding to the mining unit is the push service portrait The running configuration information contained in the node is in addition to the running configuration information of the currently configured subscribed push group.

子步骤S137，根据挖掘单位对应的运行配置信息，建立挖掘单位的挖掘业务关系，并确定挖掘业务关系的业务匹配元素，得到业务匹配元素中第一挖掘单位对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘的初步挖掘信息。Sub-step S137, according to the operation configuration information corresponding to the mining unit, establish the mining business relationship of the mining unit, and determine the business matching element of the mining business relationship, and obtain the knowledge map of each subscribed push group by the first mining unit in the business matching element Preliminary mining information for data collection for big data mining.

子步骤S138，在按照挖掘单位的层级依次对第一挖掘单位之后的每一挖掘单位进行初步挖掘信息筛选时，对该挖掘单位及该挖掘单位之后的每一挖掘单位的初步挖掘信息进行筛选，根据筛选后的初步挖掘信息，重新建立挖掘单位的挖掘业务关系，确定重新建立的挖掘业务关系的业务匹配元素，得到该重新建立的挖掘业务关系的业务匹配元素中该挖掘单位的筛选初步挖掘信息。Sub-step S138, when performing preliminary mining information screening for each mining unit after the first mining unit in turn according to the level of the mining unit, screening the preliminary mining information of the mining unit and each mining unit after the mining unit, According to the filtered preliminary mining information, re-establish the mining business relationship of the mining unit, determine the business matching element of the re-established mining business relationship, and obtain the preliminary mining information of the mining unit in the business matching element of the re-established mining business relationship .

子步骤S139，在得到所有挖掘单位的筛选初步挖掘信息后，将所有挖掘单位的筛选初步挖掘信息作为大数据挖掘结果。In sub-step S139, after obtaining the screening preliminary mining information of all mining units, the screening preliminary mining information of all mining units is used as the big data mining result.

在一种可能的实现方式中，在步骤S130之后，还可以包括以下步骤：In a possible implementation manner, after step S130, the following steps may also be included:

步骤S140，在大数据挖掘过程中判断是否存在用于表示可挖掘目标服务存在扩展加载业务的扩展加载业务信息，并在检测到扩展加载业务信息时，提取大数据挖掘的扩展加载业务信息对应的第一可挖掘目标服务的第一知识图谱以及与第一可挖掘目标服务存在扩展加载业务关系的至少一个第二可挖掘目标服务的第二知识图谱。Step S140, in the big data mining process, it is judged whether there is extended loading business information used to indicate that the extensible loading business exists in the mineable target service, and when the extended loading business information is detected, extract the corresponding data of the extended loading business information of big data mining. The first knowledge graph of the first mineable target service and the second knowledge graph of at least one second mineable target service that has an extended loading business relationship with the first mineable target service.

步骤S150，根据预设人工智能模型确定第一知识图谱和至少一个第二知识图谱之间的全局大数据挖掘信息。Step S150: Determine global big data mining information between the first knowledge graph and at least one second knowledge graph according to a preset artificial intelligence model.

在一种可能的实现方式中，针对步骤S140，可以从大数据挖掘过程中产生的大数据挖掘记录信息中提取大数据挖掘的扩展加载业务信息对应的第一可挖掘目标服务的第一知识图谱以及与第一可挖掘目标服务存在扩展加载业务关系的至少一个第二可挖掘目标服务的第二知识图谱。其中，与第一可挖掘目标服务存在扩展加载业务关系的至少一个第二可挖掘目标服务可以是指与第一可挖掘目标服务存在相关联的联动效应的第二可挖掘目标服务。In a possible implementation manner, for step S140, the first knowledge graph of the first mineable target service corresponding to the extended loading business information of big data mining may be extracted from the big data mining record information generated in the big data mining process and a second knowledge graph of at least one second mineable target service that has an extended loading business relationship with the first mineable target service. The at least one second mineable target service that has an extended loading business relationship with the first mineable target service may refer to a second mineable target service that has a linkage effect associated with the first mineable target service.

例如，如果某个可挖掘目标服务需要在第一可挖掘目标服务挖掘的过程中扩展挖掘，那么该可挖掘目标服务可以理解为与第一可挖掘目标服务存在扩展加载业务关系的第二可挖掘目标服务。For example, if a certain mineable target service needs to expand mining during the mining process of the first mineable target service, the mineable target service can be understood as a second mineable target service that has an extension loading business relationship with the first mineable target service. target service.

在一种可能的实现方式中，针对步骤S150，可以通过以下示例性的子步骤实现，详细描述如下。In a possible implementation manner, step S150 may be implemented through the following exemplary sub-steps, which are described in detail as follows.

子步骤S151，将第一知识图谱按照每个相同的知识图谱节点与至少一个第二知识图谱对应的知识图谱节点进行融合后，得到融合知识图谱。Sub-step S151, after the first knowledge graph is fused with at least one knowledge graph node corresponding to the second knowledge graph according to each same knowledge graph node, a fused knowledge graph is obtained.

子步骤S152，将第一知识图谱和至少一个第二知识图谱添加到预设的数据地图分类队列，并基于数据地图分类队列建立第一知识图谱的多个第一数据地图分类参数以及第二知识图谱的多个第二数据地图分类参数。Sub-step S152, adding the first knowledge graph and at least one second knowledge graph to a preset data map classification queue, and establishing a plurality of first data map classification parameters and second knowledge of the first knowledge graph based on the data map classification queue A plurality of second data map classification parameters of the map.

子步骤S153，根据每个第一数据地图分类参数确定第一可挖掘目标服务的第一知识表达信息，并根据每个第二数据地图分类参数确定第二可挖掘目标服务的第二知识表达信息，而后将第一知识表达信息和第二知识表达信息映射至知识实体特征模型，得到第一知识表达信息对应的第一知识图谱特征以及第二知识表达信息对应的第二知识图谱特征，并确定知识实体特征模型对应于融合知识图谱的多个知识语料对象，对多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表，针对每个知识语料挖掘列表，在预设的大数据挖掘进程中挖掘知识语料挖掘列表中的每个知识语料对象对应第一知识图谱特征的第一语料画像刻画特征和对应第二知识图谱特征的第二语料画像刻画特征。Sub-step S153: Determine the first knowledge expression information of the first mineable target service according to each first data map classification parameter, and determine the second knowledge expression information of the second mineable target service according to each second data map classification parameter , and then map the first knowledge expression information and the second knowledge expression information to the knowledge entity feature model, obtain the first knowledge graph feature corresponding to the first knowledge expression information and the second knowledge graph feature corresponding to the second knowledge expression information, and determine The knowledge entity feature model corresponds to a plurality of knowledge corpus objects that fuse the knowledge graph. The multiple knowledge corpus objects are aggregated to obtain at least a plurality of knowledge corpus mining lists of different categories. For each knowledge corpus mining list, in the preset big data In the mining process, each knowledge corpus object in the mining list of knowledge corpus corresponds to the first corpus portrait feature corresponding to the first knowledge graph feature and the second corpus portrait feature corresponding to the second knowledge graph feature.

子步骤S154，根据知识语料挖掘列表中的每个知识语料对象对应的第一语料画像刻画特征和第二语料画像刻画特征的挖掘结果，按照知识预料的预设优先级进行拼接生成的模拟挖掘流，对拼接生成的模拟挖掘流进行还原，确定第一可挖掘目标服务和至少一个第二可挖掘目标服务的全局大数据挖掘信息。Sub-step S154, according to the mining results of the first corpus portrait depiction feature and the second corpus portrait depiction feature corresponding to each knowledge corpus object in the knowledge corpus mining list, splicing and generating a simulated mining flow according to a preset priority of knowledge expectations. , restore the simulated mining flow generated by the splicing, and determine the global big data mining information of the first mineable target service and the at least one second mineable target service.

如此，可以在实际大数据挖掘过程中针对性地以关联的可挖掘目标服务为独立挖掘目标进行后续的大数据挖掘。In this way, in the actual big data mining process, subsequent big data mining can be carried out with the associated minable target service as the independent mining target.

示例性地，在子步骤S152中，可以通过以下详细的实施方式实现，例如可以描述如下。Exemplarily, in sub-step S152, it can be implemented by the following detailed implementation manner, for example, it can be described as follows.

（1）确定数据地图分类队列的扩展挖掘配置信息。(1) Determine the extended mining configuration information of the data map classification queue.

本实施例中，扩展挖掘配置信息用于表征数据地图分类队列对先后添加到的知识图谱进行处理时所分配的数据地图分类目标，数据地图分类目标用于表征数据地图分类队列对添加到的知识图谱进行挖掘时的挖掘特征节点信息。In this embodiment, the extended mining configuration information is used to represent the data map classification target assigned when the data map classification queue processes the knowledge maps added to it successively, and the data map classification target is used to represent the knowledge added to the data map classification queue pair. Mining feature node information when mining the graph.

（2）基于扩展挖掘配置信息，确定将第一知识图谱添加到数据地图分类队列所对应的第一挖掘特征节点信息以及将第二知识图谱添加到数据地图分类队列所对应的第二挖掘特征节点信息。(2) Based on the extended mining configuration information, determine the first mining feature node information corresponding to adding the first knowledge graph to the data map classification queue and the second mining feature node corresponding to adding the second knowledge graph to the data map classification queue information.

（3）根据第一挖掘特征节点信息和第二挖掘特征节点信息确定在将第一知识图谱和第二知识图谱添加到数据地图分类队列时是否存在扩展加载业务。(3) According to the first mining feature node information and the second mining feature node information, it is determined whether there is an extended loading service when the first knowledge graph and the second knowledge graph are added to the data map classification queue.

本实施例中，扩展加载业务可以用于表征数据地图分类队列的挖掘存在扩展加载的响应行为。In this embodiment, the extended loading service can be used to characterize the response behavior of extended loading in the mining of the data map classification queue.

（4）若确定在将第一知识图谱和第二知识图谱添加到数据地图分类队列时不存在扩展加载业务，则对第二挖掘特征节点信息进行调整得到第三挖掘特征节点信息，并基于第一挖掘特征节点信息和第三挖掘特征节点信息将第一知识图谱和第二知识图谱添加到数据地图分类队列。(4) If it is determined that there is no extended loading service when the first knowledge graph and the second knowledge graph are added to the data map classification queue, the second mining feature node information is adjusted to obtain the third mining feature node information, and based on the third mining feature node information. The first mining feature node information and the third mining feature node information add the first knowledge graph and the second knowledge graph to the data map classification queue.

本实施例中，第三挖掘特征节点信息与第二挖掘特征节点信息之间的特征差距与第一挖掘特征节点信息和第二挖掘特征节点信息之间的特征差距匹配。In this embodiment, the feature gap between the third mining feature node information and the second mining feature node information matches the feature gap between the first mining feature node information and the second mining feature node information.

（5）若确定在将第一知识图谱和第二知识图谱添加到数据地图分类队列时存在扩展加载业务，则持续采用第一挖掘特征节点信息和第二挖掘特征节点信息将第一知识图谱和第二知识图谱添加到数据地图分类队列。(5) If it is determined that there is an extended loading service when the first knowledge graph and the second knowledge graph are added to the data map classification queue, the first knowledge graph and the second mining feature node information are continuously used to A second knowledge graph is added to the datamap classification queue.

在一种可能的实现方式中，仍旧在子步骤S152中，在基于数据地图分类队列建立第一知识图谱的多个第一数据地图分类参数以及第二知识图谱的多个第二数据地图分类参数的过程中，可以通过以下详细的实施方式实现，例如可以描述如下。In a possible implementation manner, still in sub-step S152, a plurality of first data map classification parameters of the first knowledge graph and a plurality of second data map classification parameters of the second knowledge graph are established based on the data map classification queue The process can be implemented by the following detailed implementations, for example, can be described as follows.

（6）基于数据地图分类队列确定第一知识图谱的第一挖掘节点序列以及第二知识图谱的第二挖掘节点序列。(6) Determine the first mining node sequence of the first knowledge graph and the second mining node sequence of the second knowledge graph based on the data map classification queue.

其中，值得说明的是，挖掘节点序列可以用于表征知识图谱在不同挖掘节点下的挖掘业务关系，譬如可以表示过渡挖掘业务关系、覆盖挖掘业务关系、增加挖掘业务关系等等，在此不作具体想定。Among them, it is worth noting that the mining node sequence can be used to represent the mining business relationship of the knowledge graph under different mining nodes, for example, it can represent transition mining business relationship, coverage mining business relationship, adding mining business relationship, etc. Scenario.

（7）分别根据第一挖掘节点序列以及第二挖掘节点序列在数据地图分类队列中建立第一知识图谱的多个第一数据地图分类参数以及第二知识图谱的多个第二数据地图分类参数。(7) Establish a plurality of first data map classification parameters of the first knowledge graph and a plurality of second data map classification parameters of the second knowledge graph in the data map classification queue according to the first mining node sequence and the second mining node sequence respectively .

在一种可能的实现方式中，针对步骤S153，为了保证同步性和连贯性，便于后续分析，可以通过以下详细的实施方式实现，例如可以描述如下。In a possible implementation manner, for step S153, in order to ensure synchronization and coherence and facilitate subsequent analysis, it may be implemented through the following detailed implementation manner, for example, it may be described as follows.

（1）根据每个第一数据地图分类参数中的多个挖掘节点以及每相邻两个挖掘节点之间的挖掘画像地图参数确定每个第一数据地图分类参数对应的挖掘节点业务位图(1) Determine the mining node business bitmap corresponding to each first data map classification parameter according to the multiple mining nodes in each first data map classification parameter and the mining portrait map parameters between every two adjacent mining nodes

（2）基于挖掘节点业务位图确定第一可挖掘目标服务的第一知识表达信息。(2) Determine the first knowledge expression information of the first mineable target service based on the mining node service bitmap.

其中，第一数据地图分类参数中的每个挖掘节点对应设置有挖掘画像地图索引参数，挖掘画像地图索引参数与任意一个挖掘节点的挖掘画像地图索引参数之间的匹配参数作为对应的挖掘画像地图参数，挖掘画像地图索引参数根据挖掘节点在第一数据地图分类参数中的挖掘频繁项模式确定。Wherein, each mining node in the first data map classification parameter is correspondingly set with a mining portrait map index parameter, and a matching parameter between the mining portrait map index parameter and the mining portrait map index parameter of any mining node is used as the corresponding mining portrait map parameter, the mining portrait map index parameter is determined according to the mining frequent item mode of the mining node in the first data map classification parameter.

（3）将每个第二数据地图分类参数的挖掘节点和挖掘节点对应的挖掘画像地图索引参数列出，得到每个第二数据地图分类参数对应的第一定位知识意图和第二定位知识意图。(3) List the mining nodes of each second data map classification parameter and the mining portrait map index parameters corresponding to the mining nodes, and obtain the first positioning knowledge intention and the second positioning knowledge intention corresponding to each second data map classification parameter .

其中，示例性地，第一定位知识意图可以为第二数据地图分类参数的挖掘节点对应的定位知识意图，第二定位知识意图可以为第二数据地图分类参数的挖掘画像地图索引参数对应的定位知识意图。Wherein, for example, the first location knowledge intent may be the location knowledge intent corresponding to the mining node of the second data map classification parameter, and the second location knowledge intent may be the location corresponding to the mining portrait map index parameter of the second data map classification parameter Knowledge intent.

（4）确定第一定位知识意图相对于第二定位知识意图的第一意图抽取实体以及第二定位知识意图相对于第二定位知识意图的第二意图抽取实体。(4) Determine the first intent extraction entity of the first location knowledge intent relative to the second location knowledge intent and the second intent extraction entity of the second location knowledge intent relative to the second location knowledge intent.

（5）获取第一意图抽取实体和第二意图抽取实体中具有相同的抽取实体连续性的至少三个目标抽取实体节点，并根据目标抽取实体节点确定出第二数据地图分类参数的第二知识表达信息。(5) Obtain at least three target extraction entity nodes that have the same extraction entity continuity in the first intent extraction entity and the second intent extraction entity, and determine the second knowledge of the second data map classification parameter according to the target extraction entity node express information.

其中，示例性地，抽取实体连续性用于表征每两个抽取实体之间的实体循环关系。Wherein, exemplarily, the extracted entity continuity is used to represent the entity cyclic relationship between every two extracted entities.

在一种可能的实现方式中，仍旧针对步骤S153，在对多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表的过程中，可以通过以下详细的实施方式实现，例如可以描述如下。In a possible implementation manner, still for step S153, in the process of aggregating multiple knowledge corpus objects to obtain at least multiple knowledge corpus mining lists of different categories, the following detailed implementation can be used to implement, for example, it can be described as follows.

（6）确定知识实体特征模型中的每个知识语料对象对应的知识图谱特征的业务标签范围。(6) Determine the business label range of the knowledge graph feature corresponding to each knowledge corpus object in the knowledge entity feature model.

（7）确定每个知识语料对象对应的知识图谱特征的图谱节点重合范围。(7) Determine the overlapping range of the graph nodes of the knowledge graph feature corresponding to each knowledge corpus object.

其中，图谱节点重合范围可以为每个知识语料对象对应的知识图谱特征中第一知识图谱特征与第二知识图谱特征的重合部分。The overlapping range of the graph nodes may be the overlapping portion of the first knowledge graph feature and the second knowledge graph feature in the knowledge graph feature corresponding to each knowledge corpus object.

（8）确定每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的融合挖掘信息。(8) Determine the fusion mining information of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object.

其中，融合挖掘信息可以通过对第一知识图谱特征和第二知识图谱特征对应设定业务标签范围的挖掘指向对象进行特征参数并集计算得到。Wherein, the fusion mining information can be obtained by performing feature parameter union calculation on the mining pointing objects whose business label ranges are set corresponding to the first knowledge map feature and the second knowledge map feature.

（9）根据每个知识语料对象对应的知识图谱特征的业务标签范围、图谱节点重合范围和融合挖掘信息确定每个知识语料对象的结构化主题特征序列（即以知识图谱特征的业务标签范围、图谱节点重合范围和融合挖掘信息为顺序形成的序列）。(9) Determine the structured topic feature sequence of each knowledge corpus object according to the business label range of the knowledge graph feature corresponding to each knowledge corpus object, the overlapping range of graph nodes and the fusion mining information (that is, the business label range of the knowledge graph feature, The overlapping range of the graph nodes and the fusion mining information are sequentially formed sequences).

（10）基于每个知识语料对象的结构化主题特征序列对每个知识语料对象进行汇总，得到至少多个不同类别的知识语料挖掘列表。(10) Summarize each knowledge corpus object based on the structured topic feature sequence of each knowledge corpus object, and obtain at least multiple knowledge corpus mining lists of different categories.

例如，可以把结构化主题特征序列中的各个特征参数存在至少一个相同特征参数的知识语料对象汇总到该相同特征参数所对应的类别的知识语料挖掘列表，从而到至少多个不同类别的知识语料挖掘列表。For example, each feature parameter in the structured topic feature sequence can be aggregated into the knowledge corpus mining list of the category corresponding to the same feature parameter with at least one knowledge corpus object of the same feature parameter, so as to obtain at least a plurality of different categories of knowledge corpus Dig the list.

在一种可能的实现方式中，仍旧针对步骤S153，在预设的大数据挖掘进程中挖掘知识语料挖掘列表中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的过程中，可以通过以下详细的实施方式实现，例如可以描述如下。In a possible implementation manner, still for step S153, the process of mining the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object in the knowledge corpus mining list in the preset big data mining process can be realized by the following detailed embodiments, for example, can be described as follows.

（11）确定每个知识语料挖掘列表中每个知识语料对象对应的结构化主题特征序列的扩展挖掘配置信息。(11) Determine the extended mining configuration information of the structured topic feature sequence corresponding to each knowledge corpus object in each knowledge corpus mining list.

（12）根据扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差。(12) Determine the data map classification error of the first knowledge map feature and the second knowledge map feature corresponding to each knowledge corpus object in each summary according to the extended mining configuration information.

其中，数据地图分类误差可以用于表征每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的挖掘误差情况。The data map classification error can be used to represent the mining error of the first knowledge map feature and the second knowledge map feature corresponding to each knowledge corpus object.

（13）判断每个数据地图分类误差与大数据挖掘进程对应的基准挖掘误差的差值是否在预设差值区间内。(13) Determine whether the difference between the classification error of each data map and the benchmark mining error corresponding to the big data mining process is within a preset difference interval.

其中，预设差值区间可以用于表征大数据挖掘进程处于正常运行时每个数据地图分类误差所处的区间。The preset difference interval can be used to represent the interval in which the classification error of each data map is located when the big data mining process is in normal operation.

（13）在每个数据地图分类误差与大数据挖掘进程对应的基准同步系数的差值均落入预设差值区间时，可以基于大数据挖掘进程模拟验证知识语料挖掘列表中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征。(13) When the difference between the classification error of each data map and the reference synchronization coefficient corresponding to the big data mining process falls within the preset difference range, each knowledge in the knowledge corpus mining list can be simulated and verified based on the big data mining process. The first knowledge graph feature and the second knowledge graph feature corresponding to the corpus object.

（14）否则，在每个数据地图分类误差与大数据挖掘进程对应的基准同步系数的差值均未落入预设差值区间时，可以根据大数据挖掘进程的参数更新子进程对未落入预设差值区间内的差值对应的数据地图分类误差对应的扩展挖掘配置信息进行修正，并返回根据扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差的步骤。(14) Otherwise, when the difference between the classification error of each data map and the reference synchronization coefficient corresponding to the big data mining process does not fall within the preset difference range, the sub-process can be updated according to the parameters of the big data mining process. The extended mining configuration information corresponding to the data map classification error corresponding to the difference within the preset difference interval is corrected, and the first knowledge map feature corresponding to each knowledge corpus object in each summary is determined according to the extended mining configuration information. and the second knowledge graph feature data map classification error step.

在一种可能的实现方式中，针对步骤S154，譬如，在对拼接生成的模拟挖掘流进行还原，确定第一可挖掘目标服务和至少一个第二可挖掘目标服务的全局大数据挖掘信息的过程中，可以对拼接生成的模拟挖掘流按照每个对应的模拟挖掘节点进行逆转换，以获得第一可挖掘目标服务和至少一个第二可挖掘目标服务的全局大数据挖掘信息。In a possible implementation manner, for step S154, for example, when restoring the simulated mining flow generated by splicing, the process of determining the global big data mining information of the first mineable target service and the at least one second mineable target service , the simulated mining flow generated by the splicing can be inversely transformed according to each corresponding simulated mining node, so as to obtain the global big data mining information of the first mineable target service and the at least one second mineable target service.

图3为本公开实施例提供的基于区块链离线支付的大数据处理装置300的功能模块示意图，本实施例可以根据上述云服务推送平台100执行的方法实施例对该基于区块链离线支付的大数据处理装置300进行功能模块的划分，也即该基于区块链离线支付的大数据处理装置300所对应的以下各个功能模块可以用于执行上述云服务推送平台100执行的各个方法实施例。其中，该基于区块链离线支付的大数据处理装置300可以包括获取模块310、分组模块320以及大数据挖掘模块330，下面分别对该基于区块链离线支付的大数据处理装置300的各个功能模块的功能进行详细阐述。FIG. 3 is a schematic diagram of functional modules of a bigdata processing apparatus 300 based on blockchain offline payment according to an embodiment of the present disclosure. In this embodiment, the blockchain-based offline payment can be performed according to the method embodiment performed by the cloudservice push platform 100 described above. The bigdata processing apparatus 300 according to the above-mentioned bigdata processing apparatus 300 divides the functional modules, that is, the following functional modules corresponding to the bigdata processing apparatus 300 based on blockchain offline payment can be used to execute the various method embodiments executed by the cloudservice push platform 100. . The bigdata processing device 300 based on blockchain offline payment may include anacquisition module 310, agrouping module 320, and a bigdata mining module 330. The following are the functions of the bigdata processing device 300 based on blockchain offline payment. The function of the module is explained in detail.

获取模块310，用于从每个数字金融服务终端200中获取每个数字金融服务终端200在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素。其中，获取模块310可以用于执行上述的步骤S110，关于获取模块310的详细实现方式可以参照上述针对步骤S110的详细描述即可。The acquiringmodule 310 is configured to acquire, from each digitalfinancial service terminal 200, the offline bill data set generated by each digitalfinancial service terminal 200 under the blockchain offline payment environment and the target payment environment corresponding to the offline bill data set element. The obtainingmodule 310 may be configured to execute the above-mentioned step S110, and for the detailed implementation of the obtainingmodule 310, reference may be made to the above-mentioned detailed description of the step S110.

分组模块320，用于获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合。其中，分组模块320可以用于执行上述的步骤S120，关于分组模块320的详细实现方式可以参照上述针对步骤S120的详细描述即可。Thegrouping module 320 is configured to obtain the minable target services under the target payment environment element with the service tag to be mined, and group the minable target services under each target payment environment element according to a predetermined subscribed push group, and generate A collection of minable target services for each subscribed push group. Thegrouping module 320 may be configured to perform the above-mentioned step S120, and for the detailed implementation of thegrouping module 320, reference may be made to the above-mentioned detailed description of the step S120.

大数据挖掘模块330，用于针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。其中，大数据挖掘模块330可以用于执行上述的步骤S130，关于大数据挖掘模块330的详细实现方式可以参照上述针对步骤S130的详细描述即可。The bigdata mining module 330 is configured to, for each subscribed push group, obtain the knowledge graph data of each minable target service in the minable target service set of the subscribed push group that matches the offline billing data set, and based on The push service portrait corresponding to the subscribed push group performs big data mining on the knowledge graph data set of each subscribed push group. The bigdata mining module 330 may be used to execute the above-mentioned step S130, and for the detailed implementation of the bigdata mining module 330, refer to the above-mentioned detailed description of the step S130.

需要说明的是，应理解以上装置的各个模块的划分仅仅是一种逻辑功能的划分，实际实现时可以全部或部分集成到一个物理实体上，也可以物理上分开。且这些模块可以全部以软件通过处理元件调用的形式实现。也可以全部以硬件的形式实现。还可以部分模块通过处理元件调用软件的形式实现，部分模块通过硬件的形式实现。例如，获取模块310可以为单独设立的处理元件，也可以集成在上述装置的某一个芯片中实现，此外，也可以以程序代码的形式存储于上述装置的存储器中，由上述装置的某一个处理元件调用并执行以上获取模块310的功能。其它模块的实现与之类似。此外这些模块全部或部分可以集成在一起，也可以独立实现。这里所描述的处理元件可以是一种集成电路，具有信号的处理能力。在实现过程中，上述方法的各步骤或以上各个模块可以通过处理器元件中的硬件的集成逻辑电路或者软件形式的指令完成。It should be noted that it should be understood that the division of each module of the above apparatus is only a division of logical functions, and may be fully or partially integrated into a physical entity in actual implementation, or may be physically separated. And these modules can all be implemented in the form of software calling through processing elements. All can also be implemented in the form of hardware. Some modules can also be implemented in the form of calling software through processing elements, and some modules can be implemented in the form of hardware. For example, theacquisition module 310 may be a separately established processing element, or may be integrated into a certain chip of the above-mentioned apparatus to realize, in addition, it may also be stored in the memory of the above-mentioned apparatus in the form of program code, and processed by one of the above-mentioned apparatuses The element invokes and executes the functions of theacquisition module 310 above. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, and can also be implemented independently. The processing element described herein may be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above-mentioned method or each of the above-mentioned modules can be completed by an integrated logic circuit of hardware in the processor element or an instruction in the form of software.

例如，以上这些模块可以是被配置成实施以上方法的一个或多个集成电路，例如：一个或多个特定集成电路(application specific integrated circuit，ASIC)，或，一个或多个微处理器(digital signal processor，DSP)，或，一个或者多个现场可编程门阵列(field programmable gate array，FPGA)等。再如，当以上某个模块通过处理元件调度程序代码的形式实现时，该处理元件可以是通用处理器，例如中央处理器(centralprocessing unit，CPU)或其它可以调用程序代码的处理器。再如，这些模块可以集成在一起，以片上系统(system-on-a-chip，SOC)的形式实现。For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as: one or more application specific integrated circuits (ASIC), or one or more digital microprocessors (digital) signal processor, DSP), or, one or more field programmable gate array (field programmable gate array, FPGA) and so on. For another example, when one of the above modules is implemented in the form of a processing element scheduling program code, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processors that can invoke program codes. For another example, these modules can be integrated together and implemented in the form of a system-on-a-chip (SOC).

图4示出了本公开实施例提供的用于实现上述的控制设备的云服务推送平台100的硬件结构示意图，如图4所示，云服务推送平台100可包括处理器110、机器可读存储介质120、总线130以及收发器140。FIG. 4 shows a schematic diagram of the hardware structure of the cloudservice push platform 100 provided by an embodiment of the present disclosure for implementing the above-mentioned control device. As shown in FIG. 4 , the cloudservice push platform 100 may include aprocessor 110, a machine-readable storage Media 120 ,bus 130 andtransceiver 140 .

在具体实现过程中，至少一个处理器110执行机器可读存储介质120存储的计算机执行指令（例如图3中所示的基于区块链离线支付的大数据处理装置300包括的获取模块310、分组模块320以及大数据挖掘模块330），使得处理器110可以执行如上方法实施例的基于区块链离线支付的大数据处理方法，其中，处理器110、机器可读存储介质120以及收发器140通过总线130连接，处理器110可以用于控制收发器140的收发动作，从而可以与前述的数字金融服务终端200进行数据收发。In a specific implementation process, at least oneprocessor 110 executes computer-executable instructions stored in the machine-readable storage medium 120 (for example, theacquisition module 310,group module 320 and big data mining module 330), so that theprocessor 110 can execute the blockchain offline payment-based big data processing method in the above method embodiment, wherein theprocessor 110, the machine-readable storage medium 120 and thetransceiver 140 pass through Thebus 130 is connected, and theprocessor 110 can be used to control thetransceiver 140 to transmit and receive data, so as to transmit and receive data with the aforementioned digitalfinancial service terminal 200 .

处理器110的具体实现过程可参见上述云服务推送平台100执行的各个方法实施例，其实现原理和技术效果类似，本实施例此处不再赘述。For the specific implementation process of theprocessor 110, reference may be made to the various method embodiments executed by the cloudservice push platform 100 above, and the implementation principles and technical effects thereof are similar, and details are not described herein again in this embodiment.

在上述的图4所示的实施例中，应理解，处理器可以是中央处理单元(英文：Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(英文：Digital Signal Processor，DSP)、专用集成电路(英文：ApplicationSpecificIntegrated Circuit，ASIC)等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等。结合发明所公开的方法的步骤可以直接体现为硬件处理器执行完成，或者用处理器中的硬件及软件模块组合执行完成。In the above-mentioned embodiment shown in FIG. 4 , it should be understood that the processor may be a central processing unit (English: Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (English: Digital Signal Processor) , DSP), application specific integrated circuit (English: ApplicationSpecificIntegrated Circuit, ASIC) and so on. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in conjunction with the invention can be directly embodied as executed by a hardware processor, or executed by a combination of hardware and software modules in the processor.

机器可读存储介质120可能包含高速RAM存储器，也可能还包括非易失性存储NVM，例如至少一个磁盘存储器。Machine-readable storage medium 120 may include high-speed RAM memory, and may also include non-volatile storage NVM, such as at least one disk storage.

总线130可以是工业标准体系结构(Industry Standard Architecture，ISA)总线、外部设备互连(Peripheral Component Interconnect，PCI)总线或扩展工业标准体系结构(Extended Industry Standard Architecture，EISA)总线等。总线130可以分为地址总线、数据总线、控制总线等。为便于表示，本申请附图中的总线并不限定仅有一根总线或一种类型的总线。Thebus 130 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. Thebus 130 may be classified into an address bus, a data bus, a control bus, and the like. For convenience of representation, the buses in the drawings of the present application are not limited to only one bus or one type of bus.

此外，本公开实施例还提供一种可读存储介质，所述可读存储介质中存储有计算机执行指令，当处理器执行所述计算机执行指令时，实现如上基于区块链离线支付的大数据处理方法。In addition, an embodiment of the present disclosure also provides a readable storage medium, where computer-executable instructions are stored in the readable storage medium, and when a processor executes the computer-executable instructions, the above-mentioned big data for offline payment based on blockchain is realized. Approach.

上述的可读存储介质可以是由任何类型的易失性或非易失性存储设备或者它们的组合实现，如静态随机存取存储器(SRAM)，电可擦除可编程只读存储器(EEPROM)，可擦除可编程只读存储器(EPROM)，可编程只读存储器(PROM)，只读存储器(ROM)，磁存储器，快闪存储器，磁盘或光盘。可读存储介质可以是通用或专用计算机能够存取的任何可用介质。The above-mentioned readable storage medium may be implemented by any type of volatile or non-volatile storage device or combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM) , Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic Disk or Optical Disk. A readable storage medium can be any available medium that can be accessed by a general purpose or special purpose computer.

最后应说明的是：以上各实施例仅用以说明本公开的技术方案，而非对其限制；尽管参照前述各实施例对本公开进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本公开各实施例技术方案的范围。Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, but not to limit them; although the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features thereof can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present disclosure. scope.

Claims (10)

Translated fromChinese

1.一种基于区块链离线支付的大数据处理方法，其特征在于，应用于云服务推送平台，所述云服务推送平台与多个数字金融服务终端通信连接，所述方法包括：1. a big data processing method based on blockchain offline payment, is characterized in that, is applied to cloud service push platform, and described cloud service push platform is connected with multiple digital financial service terminals in communication, and described method comprises:从每个数字金融服务终端中获取每个数字金融服务终端在区块链离线支付环境下生成的离线账单数据集合以及所述离线账单数据集合所对应的目标支付环境元素；Obtain from each digital financial service terminal the offline bill data set generated by each digital financial service terminal in the blockchain offline payment environment and the target payment environment element corresponding to the offline bill data set;获取待挖掘服务标签在所述目标支付环境元素下的可挖掘目标服务，并按照预定的已订阅推送分组对各个目标支付环境元素下的可挖掘目标服务进行分组，分别生成每个已订阅推送分组的可挖掘目标服务集合；Acquire the minable target services with the service tag to be mined under the target payment environment element, and group the mineable target services under each target payment environment element according to the predetermined subscribed push grouping, and generate each subscribed push grouping respectively A collection of minable target services;针对每个已订阅推送分组，获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据，并基于已订阅推送分组对应的推送服务画像对每个已订阅推送分组的知识图谱数据集合进行大数据挖掘。For each subscribed push group, acquire the knowledge graph data of each minable target service in the subscribed push group's minable target service set that matches the offline billing data set, and based on the push service corresponding to the subscribed push group The portrait performs big data mining on the knowledge graph data set of each subscribed push group.2.根据权利要求1所述的基于区块链离线支付的大数据处理方法，其特征在于，所述获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据的步骤，包括：2. The big data processing method based on blockchain offline payment according to claim 1, wherein each minable target service in the minable target service set of the subscribed push group is obtained and matched with the The steps of the knowledge graph data of the offline billing data collection include:获取该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务相关的匹配关键词向量；Obtain the matching keyword vector related to each minable target service in the set of minable target services of the subscribed push group;根据所述每个可挖掘目标服务相关的匹配关键词向量从所述离线账单数据集合中匹配对应的账单板块内容；Match the corresponding billing section content from the offline billing data set according to the matching keyword vector related to each mineable target service;根据所述每个可挖掘目标服务相关的匹配关键词向量匹配的账单板块内容中每个业务记录板块对应的知识图谱内容，确定该已订阅推送分组的可挖掘目标服务集合中每个可挖掘目标服务匹配于所述离线账单数据集合的知识图谱数据。According to the knowledge graph content corresponding to each business record section in the billing section content matched by the matching keyword vector related to each minable target service, determine each minable target in the set of minable target services of the subscribed push group The service matches the knowledge graph data of the offline billing data set.3.根据权利要求1或2所述的基于区块链离线支付的大数据处理方法，其特征在于，所述方法还包括：3. The big data processing method based on blockchain offline payment according to claim 1 or 2, wherein the method further comprises:在大数据挖掘过程中判断是否存在用于表示可挖掘目标服务存在扩展加载业务的扩展加载业务信息，并在检测到所述扩展加载业务信息时，提取大数据挖掘的所述扩展加载业务信息对应的第一可挖掘目标服务的第一知识图谱以及与所述第一可挖掘目标服务存在扩展加载业务关系的至少一个第二可挖掘目标服务的第二知识图谱；In the process of big data mining, it is judged whether there is extended loading service information indicating that the target service can be mined and the extended loading service exists, and when the extended loading service information is detected, the corresponding information of the extended loading service information of big data mining is extracted. the first knowledge graph of the first mineable target service and the second knowledge graph of at least one second mineable target service that has an extended loading business relationship with the first mineable target service;根据预设人工智能模型确定所述第一知识图谱和至少一个第二知识图谱之间的全局大数据挖掘信息。Global big data mining information between the first knowledge graph and at least one second knowledge graph is determined according to a preset artificial intelligence model.4.根据权利要求1所述的基于区块链离线支付的大数据处理方法，其特征在于，所述根据预设人工智能模型确定所述第一知识图谱和至少一个第二知识图谱之间的全局大数据挖掘信息的步骤，包括：4. The big data processing method based on blockchain offline payment according to claim 1, characterized in that, determining the relationship between the first knowledge graph and at least one second knowledge graph according to a preset artificial intelligence model. The steps of global big data mining information include:将所述第一知识图谱按照每个相同的知识图谱节点与至少一个第二知识图谱对应的知识图谱节点进行融合后，得到融合知识图谱；After the first knowledge graph is fused with at least one knowledge graph node corresponding to the second knowledge graph according to each identical knowledge graph node, a fused knowledge graph is obtained;将所述第一知识图谱和至少一个第二知识图谱添加到预设的数据地图分类队列，并基于所述数据地图分类队列建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数；The first knowledge map and at least one second knowledge map are added to a preset data map classification queue, and a plurality of first data map classification parameters and all the first data map classification parameters of the first knowledge map are established based on the data map classification queue. multiple second data map classification parameters of the second knowledge map;根据每个第一数据地图分类参数确定所述第一可挖掘目标服务的第一知识表达信息，并根据每个第二数据地图分类参数确定所述第二可挖掘目标服务的第二知识表达信息，而后将所述第一知识表达信息和所述第二知识表达信息映射至知识实体特征模型，得到所述第一知识表达信息对应的第一知识图谱特征以及所述第二知识表达信息对应的第二知识图谱特征，并确定所述知识实体特征模型对应于所述融合知识图谱的多个知识语料对象，对所述多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表，针对每个知识语料挖掘列表，在预设的大数据挖掘进程中挖掘所述知识语料挖掘列表中的每个知识语料对象对应所述第一知识图谱特征的第一语料画像刻画特征和对应所述第二知识图谱特征的第二语料画像刻画特征；The first knowledge expression information of the first mineable target service is determined according to each first data map classification parameter, and the second knowledge expression information of the second mineable target service is determined according to each second data map classification parameter , and then map the first knowledge expression information and the second knowledge expression information to the knowledge entity feature model, and obtain the first knowledge map feature corresponding to the first knowledge expression information and the corresponding second knowledge expression information. the second knowledge graph feature, and determining that the knowledge entity feature model corresponds to a plurality of knowledge corpus objects of the fused knowledge graph, and summarizing the plurality of knowledge corpus objects to obtain at least a plurality of knowledge corpus mining lists of different categories, For each knowledge corpus mining list, in a preset big data mining process, mine each knowledge corpus object in the knowledge corpus mining list corresponding to the first corpus portrait feature of the first knowledge map feature and the corresponding The second corpus portrait of the second knowledge graph feature depicts the feature;根据所述知识语料挖掘列表中的每个知识语料对象对应的第一语料画像刻画特征和第二语料画像刻画特征的挖掘结果，按照知识预料的预设优先级进行拼接生成的模拟挖掘流，对所述拼接生成的模拟挖掘流进行还原，确定所述第一可挖掘目标服务和至少一个第二可挖掘目标服务的全局大数据挖掘信息。According to the mining results of the first corpus portrait description feature and the second corpus portrait description feature corresponding to each knowledge corpus object in the knowledge corpus mining list, the simulated mining flow generated by splicing is performed according to the preset priority of knowledge expectation, The simulated mining flow generated by the splicing is restored, and the global big data mining information of the first mineable target service and the at least one second mineable target service is determined.5.根据权利要求4所述的基于区块链离线支付的大数据处理方法，其特征在于，所述将所述第一知识图谱和至少一个第二知识图谱添加到预设的数据地图分类队列的步骤，包括：5. The big data processing method based on blockchain offline payment according to claim 4, wherein the said first knowledge graph and at least one second knowledge graph are added to a preset data map classification queue steps, including:确定所述数据地图分类队列的扩展挖掘配置信息；其中，所述扩展挖掘配置信息用于表征所述数据地图分类队列对先后添加到的知识图谱进行处理时所分配的数据地图分类目标，所述数据地图分类目标用于表征所述数据地图分类队列对添加到的知识图谱进行挖掘时的挖掘特征节点信息；Determine the extended mining configuration information of the data map classification queue; wherein, the extended mining configuration information is used to represent the data map classification target allocated when the data map classification queue processes the knowledge maps added successively, the The data map classification target is used to represent the mining feature node information when the data map classification queue mines the added knowledge map;基于所述扩展挖掘配置信息，确定将所述第一知识图谱添加到所述数据地图分类队列所对应的第一挖掘特征节点信息以及将所述第二知识图谱添加到所述数据地图分类队列所对应的第二挖掘特征节点信息；Based on the extended mining configuration information, it is determined that adding the first knowledge graph to the first mining feature node information corresponding to the data map classification queue and adding the second knowledge graph to the data map classification queue Corresponding second mining feature node information;根据所述第一挖掘特征节点信息和所述第二挖掘特征节点信息确定在将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列时是否存在扩展加载业务；其中，所述扩展加载业务用于表征所述数据地图分类队列的挖掘存在扩展加载的响应行为；Determine, according to the first mining feature node information and the second mining feature node information, whether there is an extended loading service when the first knowledge graph and the second knowledge graph are added to the data map classification queue; wherein , the extended loading service is used to characterize that the mining of the data map classification queue has a response behavior of extended loading;若否，则对所述第二挖掘特征节点信息进行调整得到第三挖掘特征节点信息，并基于所述第一挖掘特征节点信息和所述第三挖掘特征节点信息将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列，其中，所述第三挖掘特征节点信息与所述第二挖掘特征节点信息之间的特征差距与所述第一挖掘特征节点信息和所述第二挖掘特征节点信息之间的特征差距匹配；If not, the second mining feature node information is adjusted to obtain third mining feature node information, and the first knowledge graph and the third mining feature node information are combined based on the first mining feature node information and the third mining feature node information. The second knowledge graph is added to the data map classification queue, wherein the feature gap between the third mining feature node information and the second mining feature node information is the same as the first mining feature node information and all Describe the feature gap matching between the second mining feature node information;若是，则持续采用所述第一挖掘特征节点信息和所述第二挖掘特征节点信息将所述第一知识图谱和所述第二知识图谱添加到所述数据地图分类队列。If so, the first knowledge graph and the second knowledge graph are added to the data map classification queue by continuously using the first mining feature node information and the second mining feature node information.6.根据权利要求4所述的基于区块链离线支付的大数据处理方法，其特征在于，基于所述数据地图分类队列建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数的步骤，包括：6 . The big data processing method based on blockchain offline payment according to claim 4 , wherein a plurality of first data map classification parameters of the first knowledge map and all the first data map classification parameters are established based on the data map classification queue. 7 . The steps of describing a plurality of second data map classification parameters of the second knowledge map include:基于所述数据地图分类队列确定所述第一知识图谱的第一挖掘节点序列以及所述第二知识图谱的第二挖掘节点序列；其中，所述挖掘节点序列用于表征知识图谱在不同挖掘节点下的挖掘业务关系；The first mining node sequence of the first knowledge graph and the second mining node sequence of the second knowledge graph are determined based on the data map classification queue; wherein the mining node sequence is used to represent the knowledge graph at different mining nodes Mining business relationships under the分别根据所述第一挖掘节点序列以及所述第二挖掘节点序列在所述数据地图分类队列中建立所述第一知识图谱的多个第一数据地图分类参数以及所述第二知识图谱的多个第二数据地图分类参数。Establish a plurality of first data map classification parameters of the first knowledge graph and multiple data map classification parameters of the second knowledge graph in the data map classification queue according to the first mining node sequence and the second mining node sequence, respectively. A second data map classification parameter.7.根据权利要求4所述的基于区块链离线支付的大数据处理方法，其特征在于，所述根据每个第一数据地图分类参数确定所述第一可挖掘目标服务的第一知识表达信息，并根据每个第二数据地图分类参数确定所述第二可挖掘目标服务的第二知识表达信息的步骤，包括：7 . The big data processing method based on blockchain offline payment according to claim 4 , wherein the first knowledge expression of the first mineable target service is determined according to each first data map classification parameter. 8 . The step of determining the second knowledge expression information of the second mineable target service according to each second data map classification parameter includes:根据每个第一数据地图分类参数中的多个挖掘节点以及每相邻两个挖掘节点之间的挖掘画像地图参数确定每个第一数据地图分类参数对应的挖掘节点业务位图；Determine the mining node business bitmap corresponding to each first data map classification parameter according to the plurality of mining nodes in each first data map classification parameter and the mining portrait map parameters between every two adjacent mining nodes;基于所述挖掘节点业务位图确定所述第一可挖掘目标服务的第一知识表达信息；其中，所述第一数据地图分类参数中的每个挖掘节点对应设置有挖掘画像地图索引参数，所述挖掘画像地图索引参数与任意一个挖掘节点的挖掘画像地图索引参数之间的匹配参数作为对应的挖掘画像地图参数，所述挖掘画像地图索引参数根据所述挖掘节点在所述第一数据地图分类参数中的挖掘频繁项模式确定；Determine the first knowledge expression information of the first mineable target service based on the mining node business bitmap; wherein, each mining node in the first data map classification parameter is correspondingly set with a mining portrait map index parameter, so The matching parameter between the mining portrait map index parameter and the mining portrait map index parameter of any mining node is used as the corresponding mining portrait map parameter, and the mining portrait map index parameter is classified according to the mining node in the first data map The mining frequent item pattern in the parameter is determined;将每个第二数据地图分类参数的挖掘节点和挖掘节点对应的挖掘画像地图索引参数列出，得到每个第二数据地图分类参数对应的第一定位知识意图和第二定位知识意图；其中，所述第一定位知识意图为第二数据地图分类参数的挖掘节点对应的定位知识意图，所述第二定位知识意图为第二数据地图分类参数的挖掘画像地图索引参数对应的定位知识意图；The mining node of each second data map classification parameter and the mining portrait map index parameter corresponding to the mining node are listed, and the first positioning knowledge intention and the second positioning knowledge intention corresponding to each second data map classification parameter are obtained; wherein, The first positioning knowledge intention is the positioning knowledge intention corresponding to the mining node of the second data map classification parameter, and the second positioning knowledge intention is the positioning knowledge intention corresponding to the mining portrait map index parameter of the second data map classification parameter;确定所述第一定位知识意图相对于所述第二定位知识意图的第一意图抽取实体以及所述第二定位知识意图相对于所述第二定位知识意图的第二意图抽取实体；determining a first intent extraction entity of the first location knowledge intent relative to the second location knowledge intent and a second intent extraction entity of the second location knowledge intent relative to the second location knowledge intent;获取所述第一意图抽取实体和所述第二意图抽取实体中具有相同的抽取实体连续性的至少三个目标抽取实体节点，并根据所述目标抽取实体节点确定出所述第二数据地图分类参数的第二知识表达信息；其中，所述抽取实体连续性用于表征每两个抽取实体之间的实体循环关系。Obtain at least three target extraction entity nodes with the same extraction entity continuity in the first intent extraction entity and the second intent extraction entity, and determine the second data map classification according to the target extraction entity nodes The second knowledge expression information of the parameter; wherein, the extracted entity continuity is used to represent the entity cycle relationship between every two extracted entities.8.根据权利要求7所述的基于区块链离线支付的大数据处理方法，其特征在于，所述对所述多个知识语料对象进行汇总得到至少多个不同类别的知识语料挖掘列表的步骤，包括：8. The big data processing method based on blockchain offline payment according to claim 7, wherein the step of summarizing the plurality of knowledge corpus objects to obtain at least a plurality of different categories of knowledge corpus mining lists ,include:确定所述知识实体特征模型中的每个知识语料对象对应的知识图谱特征的业务标签范围；Determine the business label range of the knowledge graph feature corresponding to each knowledge corpus object in the knowledge entity feature model;确定每个知识语料对象对应的知识图谱特征的图谱节点重合范围；其中，所述图谱节点重合范围为每个知识语料对象对应的知识图谱特征中第一知识图谱特征与第二知识图谱特征的重合部分；Determine the overlap range of the graph nodes of the knowledge graph feature corresponding to each knowledge corpus object; wherein, the overlap range of the graph nodes is the overlap of the first knowledge graph feature and the second knowledge graph feature in the knowledge graph feature corresponding to each knowledge corpus object part;确定每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的融合挖掘信息；其中，所述融合挖掘信息通过对第一知识图谱特征和第二知识图谱特征对应设定业务标签范围的挖掘指向对象进行特征参数并集计算得到；Determine the fusion mining information of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object; wherein, the fusion mining information is set by correspondingly setting the business label range for the first knowledge graph feature and the second knowledge graph feature The mining points to the object to calculate the feature parameter union;根据每个知识语料对象对应的知识图谱特征的业务标签范围、图谱节点重合范围和融合挖掘信息确定每个知识语料对象的结构化主题特征序列；Determine the structured topic feature sequence of each knowledge corpus object according to the business label range, graph node coincidence range and fusion mining information of the knowledge graph feature corresponding to each knowledge corpus object;基于每个知识语料对象的结构化主题特征序列对每个知识语料对象进行汇总，得到所述至少多个不同类别的知识语料挖掘列表。Summarize each knowledge corpus object based on the structured topic feature sequence of each knowledge corpus object to obtain the at least multiple knowledge corpus mining lists of different categories.9.根据权利要求8所述的基于区块链离线支付的大数据处理方法，其特征在于，所述在预设的大数据挖掘进程中挖掘所述知识语料挖掘列表中的每个知识语料对象对应所述第一知识图谱特征的第一语料画像刻画特征和对应所述第二知识图谱特征的第二语料画像刻画特征的步骤，包括：9. The big data processing method based on blockchain offline payment according to claim 8, characterized in that, mining each knowledge corpus object in the knowledge corpus mining list in a preset big data mining process The steps of characterizing the first corpus portrait feature corresponding to the first knowledge graph feature and the second corpus portrait characterizing feature corresponding to the second knowledge graph feature include:确定每个知识语料挖掘列表中每个知识语料对象对应的结构化主题特征序列的扩展挖掘配置信息；Determine the extended mining configuration information of the structured topic feature sequence corresponding to each knowledge corpus object in each knowledge corpus mining list;根据所述扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差；其中，所述数据地图分类误差用于表征每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的挖掘误差情况；The data map classification error of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object in each summary is determined according to the extended mining configuration information; wherein, the data map classification error is used to represent each Mining errors of the first knowledge graph feature and the second knowledge graph feature corresponding to the knowledge corpus object;判断每个数据地图分类误差与所述大数据挖掘进程对应的基准挖掘误差的差值是否在预设差值区间内；其中，所述预设差值区间用于表征大数据挖掘进程处于正常运行时每个数据地图分类误差所处的区间；Determine whether the difference between the classification error of each data map and the benchmark mining error corresponding to the big data mining process is within a preset difference interval; wherein, the preset difference interval is used to indicate that the big data mining process is in normal operation is the interval in which the classification error of each data map is located;在每个数据地图分类误差与所述大数据挖掘进程对应的基准同步系数的差值均落入所述预设差值区间时，基于所述大数据挖掘进程模拟验证所述知识语料挖掘列表中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征；When the difference between each data map classification error and the reference synchronization coefficient corresponding to the big data mining process falls within the preset difference interval, simulate and verify that the knowledge corpus mining list is based on the big data mining process The first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object of ;否则，根据所述大数据挖掘进程的参数更新子进程对未落入所述预设差值区间内的差值对应的数据地图分类误差对应的扩展挖掘配置信息进行修正，并返回根据所述扩展挖掘配置信息确定每个汇总中的每个知识语料对象对应的第一知识图谱特征和第二知识图谱特征的数据地图分类误差的步骤。Otherwise, according to the parameter update sub-process of the big data mining process, correct the extended mining configuration information corresponding to the data map classification error corresponding to the difference value that does not fall within the preset difference value interval, and return according to the extended mining configuration information. The step of mining the configuration information to determine the data map classification error of the first knowledge graph feature and the second knowledge graph feature corresponding to each knowledge corpus object in each summary.10.一种云服务推送平台，其特征在于，所述云服务推送平台包括处理器、机器可读存储介质和网络接口，所述机器可读存储介质、所述网络接口以及所述处理器之间通过总线系统相连，所述网络接口用于与至少一个数字金融服务终端通信连接，所述机器可读存储介质用于存储程序、指令或代码，所述处理器用于执行所述机器可读存储介质中的程序、指令或代码，以执行权利要求1-9中任意一项的基于区块链离线支付的大数据处理方法。10. A cloud service push platform, characterized in that the cloud service push platform comprises a processor, a machine-readable storage medium, and a network interface, wherein the machine-readable storage medium, the network interface, and the processor are connected. They are connected through a bus system, the network interface is used to communicate with at least one digital financial service terminal, the machine-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the machine-readable storage medium. A program, instruction or code in the medium to execute the big data processing method based on blockchain offline payment of any one of claims 1-9.

Images