CN112069406B - Information push method and cloud service push platform based on blockchain offline payment - Google Patents

Abstract

Translated fromChinese

本申请实施例提供一种基于区块链离线支付的信息推送方法及云服务推送平台，通过从数字金融服务终端发送的基于离线支付的信息推送请求获得信息推送请求的用户订阅标签信息，然后根据订阅标签项目的大数据挖掘信息，确定出订阅标签项目对应的业务推送对象列表，从而向数字金融服务终端推送业务推送对象列表对应的业务推送定向内容。如此，可以有效针对离线支付场景的离线账单数据情况对数字金融服务终端推送业务推送对象列表对应的业务推送定向内容，从而提高信息推送的覆盖度。

The embodiments of the present application provide an information push method and a cloud service push platform based on blockchain offline payment. The user subscription label information of the information push request is obtained through an offline payment-based information push request sent from a digital financial service terminal, and then according to The big data mining information of the subscription tag item is used to determine the service push object list corresponding to the subscription tag item, so as to push the service push directional content corresponding to the service push object list to the digital financial service terminal. In this way, the service push directional content corresponding to the service push object list can be pushed to the digital financial service terminal effectively according to the offline bill data situation of the offline payment scenario, thereby improving the coverage of the information push.

Description

Information pushing method based on block chain offline payment and cloud service pushing platform

Technical Field

The application relates to the technical field of big data, in particular to an information pushing method and a cloud service pushing platform based on block chain offline payment.

Background

With the development of mobile internet technology and digital currency operation, digital currency will gradually become a new dominant payment mode in the future, and not only can support online payment, but also can support offline payment in an offline network state as in the current cash transaction.

In the offline payment state, various service bill data generated in the payment process cannot be synchronized in the cloud service platform in real time, and the offline payment scene can also reflect behavior characteristics of an offline wide user, so that subsequent service push service still needs to be provided for the offline bill data condition of the offline payment scene.

Disclosure of Invention

In order to overcome at least the above disadvantages in the prior art, an object of the present application is to provide an information push method and a cloud service push platform based on block chain offline payment, which can effectively push directional content to a service corresponding to a service push object list of a digital financial service terminal push service according to an offline bill data condition in an offline payment scenario, thereby improving coverage of information push.

In a first aspect, the present application provides an information push method based on blockchain offline payment, which is applied to a cloud service push platform, where the cloud service push platform is in communication connection with a plurality of digital financial service terminals, and the method includes:

receiving an information pushing request based on offline payment sent by the digital financial service terminal, and acquiring user subscription label information of the information pushing request; wherein the user subscription tag information comprises subscription tag items;

determining a business push object list corresponding to the subscription tag item according to big data mining information of the subscription tag item, wherein the big data mining information is obtained by carrying out big data mining on an offline bill data set generated by a digital financial service terminal in a block chain offline payment environment and a target payment environment element corresponding to the offline bill data set by the cloud service push platform;

and pushing the business push directional content corresponding to the business push object list to the digital financial service terminal.

In a possible implementation manner of the first aspect, the step of determining, according to the big data mining information of the subscription tag item, a service push object list corresponding to the subscription tag item includes:

acquiring potential demand information and demand service scene information of the potential demand information from big data mining information of the subscription tag item; the demand analysis templates used by the potential demand information and the demand service scene information are first demand analysis templates;

processing the potential demand information according to the demand service scene information to generate demand extension information of the potential demand information;

performing theme-oriented feature extraction on the potential demand information and the demand extension information, and determining second theme feature information corresponding to first theme feature information corresponding to the demand extension information from the extracted current theme-oriented feature information;

performing feature fusion on the first theme feature information and the second theme feature information to obtain third theme feature information;

outputting target theme distribution information corresponding to the potential demand information according to the third theme characteristic information; and the requirement analysis template used by the target theme distribution information is a second requirement analysis template.

In a possible implementation manner of the first aspect, the step of processing the potential demand information according to the demand service scenario information to generate demand extension information of the potential demand information includes:

performing theme-oriented feature extraction on the potential demand information, performing theme entity object identification on a first theme-oriented feature corresponding to the obtained potential demand information, and obtaining a first entity object set corresponding to the potential demand information according to the identified theme entity object;

performing theme-oriented feature extraction on the demand service scene information, performing theme entity object identification on a second theme-oriented feature corresponding to the acquired demand service scene information, and acquiring a second entity object set corresponding to the demand service scene information according to the identified theme entity object;

acquiring first entity relationship information stored in the first entity object set, and converting the first entity relationship information into a corresponding first entity relationship vector;

obtaining second entity relationship information respectively stored by a plurality of entity key nodes in the second entity object set, and converting each second entity relationship information into a corresponding second entity relationship vector;

calculating a vector inner product result of each second entity relationship vector and the first entity relationship vector;

sorting the vector inner product result corresponding to each second entity relationship vector, and selecting a plurality of similar entity relationship vectors from the second entity relationship vectors according to the sorting result;

performing convolution processing on the similar entity relationship vectors to obtain convolution relationship vectors;

carrying out entity service coverage contact ratio calculation on the first entity object set and the second entity object set, and obtaining an influence factor vector according to the entity service coverage contact ratio obtained through calculation; the influence factor vector comprises a weight parameter corresponding to each entity key node in the second entity object set;

calculating the vector inner product of the convolution relationship vector and the influence factor vector, and taking the calculated result as the expansion service vector of the first entity relationship information;

adding the expanded service vector to a potential demand unit set in the potential demand information to obtain initial expanded data;

performing subject entity object identification on the initial extension data to obtain a reference subject entity object;

and obtaining the demand extension information corresponding to the potential demand information according to the first entity object set, the second entity object set and the reference subject entity object.

In a possible implementation manner of the first aspect, the step of extracting theme-oriented features of the potential demand information and the demand extension information, and determining second theme feature information corresponding to first theme feature information corresponding to the demand extension information from the extracted current theme-oriented feature information includes:

performing theme-oriented feature extraction on the potential demand information and the demand extension information to obtain current theme-oriented feature information mapped in theme-oriented features of the potential demand information and the demand extension information; the current theme-oriented feature information comprises distribution feature information of a plurality of theme feature nodes;

determining similar distribution characteristic information of the first topic characteristic information from distribution characteristic information of a plurality of topic characteristic nodes contained in the current topic-oriented characteristic information, and taking the similar distribution characteristic information as the second topic characteristic information.

In a possible implementation manner of the first aspect, the step of performing feature fusion on the first topic feature information and the second topic feature information to obtain third topic feature information includes:

inputting the first theme feature information and the second theme feature information into a preset theme distribution analysis model respectively, so that the theme distribution analysis model outputs effective theme feature information of the first theme feature information and the second theme feature information respectively to obtain first target theme feature information and second target theme feature information;

performing convolution calculation on the first target theme characteristic information to obtain first theme convolution information; performing theme-oriented feature extraction on the first target theme feature information, performing convolution calculation on the extracted theme feature information to obtain second theme convolution information, and calculating a dot product of the first theme convolution information and the second theme convolution information to obtain a first theme feature set corresponding to the first target theme feature information;

performing convolution calculation on the second target theme characteristic information to obtain third theme convolution information; performing theme-oriented feature extraction on the second target theme feature information, performing convolution calculation on the extracted theme feature information to obtain fourth theme convolution information, and calculating a dot product of the third theme convolution information and the fourth theme convolution information to obtain a second theme feature set corresponding to the second target theme feature information;

and calculating a fusion feature set of the first topic feature set and the second topic feature set, and taking the obtained fusion feature set as the third topic feature information.

In a possible implementation manner of the first aspect, the step of outputting, according to the third topic feature information, target topic distribution information corresponding to the potential demand information includes:

obtaining a plurality of topic distribution description vectors in the third topic feature information and a topic distribution analysis strategy corresponding to each topic distribution description vector, where the topic distribution description vectors include a first topic distribution description vector and a second topic distribution description vector, where the first topic distribution description vector and the second topic distribution description vector are pairs of mapping associated topic distribution description vectors;

encoding the third topic feature information to output a first basic encoding vector corresponding to each first topic distribution description vector and a target encoding vector corresponding to the third topic feature information;

calculating the association degree between the target coding vector and each first basic coding vector to obtain the coding vector proximity degree between each corresponding first theme distribution description vector and the third theme characteristic information;

identifying all description elements in each first theme distribution description vector and the corresponding business description weight of each first theme distribution description vector;

generating a theme distribution clustering node of a corresponding theme distribution description vector according to all the description elements and the service description weight;

generating a theme distribution clustering range corresponding to each theme distribution description vector according to the theme distribution analysis strategy;

obtaining a first clustering group of each topic distribution description vector by using each topic distribution clustering node corresponding to the topic distribution clustering range;

clustering the first basic coding vector corresponding to each topic distribution description vector according to the second topic distribution description vector to obtain a second clustering group of each topic distribution description vector;

determining a target topic distribution description vector in the plurality of topic distribution description vectors according to the first clustering group and the second clustering group;

and taking the first basic coding vector corresponding to the target topic distribution description vector as a second coding vector, and adding the second coding vector to the third topic feature information to output target topic distribution information corresponding to the potential demand information.

In a possible implementation manner of the first aspect, the topic distribution parsing policy includes a vector structure clustering policy on the topic distribution description vector and a feature value clustering policy corresponding to the topic distribution description vector;

the step of generating the topic distribution clustering range corresponding to each topic distribution description vector according to the topic distribution analysis strategy comprises the following steps:

carrying out characteristic value clustering on the theme distribution description vector according to the characteristic value clustering strategy to generate a theme distribution information characteristic value corresponding to the theme distribution description vector;

carrying out vector structure clustering on the theme distribution description vector according to the vector structure clustering strategy to generate a vector structure clustering result corresponding to the theme distribution description vector;

and generating a theme distribution clustering range corresponding to the theme distribution description vector according to the theme distribution information characteristic value and the vector structure clustering result.

In a possible implementation manner of the first aspect, the topic distribution description vector further includes a third topic distribution description vector;

before the step of obtaining the respective first cluster group of each topic distribution description vector by using each topic distribution cluster node corresponding to the topic distribution cluster range, the method further includes:

calculating the coding translation parameter of each reference coding object in the reference coding object list corresponding to each first basic coding vector relative to the third topic distribution description vector;

fusing all the coding translation parameters corresponding to each first basic coding vector to obtain a fused coding parameter corresponding to each first basic coding vector;

arranging all the first basic coding vectors in sequence according to the fusion coding parameters corresponding to each first basic coding vector, and determining the respective priority parameters of each first basic coding vector according to the arranged sequence of each first basic coding vector;

processing the fusion coding parameter corresponding to each first basic coding vector according to the priority parameter of each first basic coding vector to generate a weighted fusion coding parameter of each topic distribution description vector;

the step of obtaining the respective first cluster group of each topic distribution description vector by using each topic distribution cluster node corresponding to the topic distribution cluster range includes:

and clustering the weighted fusion coding parameters of each topic distribution description vector by using each topic distribution clustering node corresponding to the topic distribution clustering range to obtain a first clustering group corresponding to each topic distribution description vector.

In a possible implementation manner of the first aspect, the method further includes:

acquiring big data mining information of the subscription tag item;

the step of obtaining big data mining information of the subscription tag item includes:

acquiring an offline bill data set generated by the digital financial service terminal in a blockchain offline payment environment and a target payment environment element corresponding to the offline bill data set from the digital financial service terminal;

acquiring excavatable target services of the service tags to be excavated under the target payment environment elements, grouping the excavatable target services under each target payment environment element according to a preset subscribed push group, and respectively generating an excavatable target service set of each subscribed push group;

acquiring, for each subscribed push group, knowledge graph data of each mineable target service in the mineable target service set of the subscribed push group, which is matched with the offline bill data set, and performing big data mining on the knowledge graph data set of each subscribed push group based on a push service graph corresponding to the subscribed push group to obtain big data mining information of each subscribed push group;

and obtaining the big data mining information of the target subscribed push group included in the subscription tag item from the big data mining information of each subscribed push group.

In a second aspect, an embodiment of the present application further provides an information pushing apparatus based on blockchain offline payment, which is applied to a cloud service pushing platform, where the cloud service pushing platform is in communication connection with a plurality of digital financial service terminals, and the apparatus includes:

the receiving module is used for receiving an information pushing request based on offline payment sent by the digital financial service terminal and obtaining user subscription label information of the information pushing request; wherein the user subscription tag information comprises subscription tag items;

the determining module is used for determining a business push object list corresponding to the subscription tag item according to big data mining information of the subscription tag item, wherein the big data mining information is obtained by carrying out big data mining on an offline bill data set generated by a digital financial service terminal in a block chain offline payment environment and a target payment environment element corresponding to the offline bill data set by the cloud service push platform;

and the pushing module is used for pushing the business pushing directional content corresponding to the business pushing object list to the digital financial service terminal.

In a third aspect, an embodiment of the present application further provides an information pushing system based on blockchain offline payment, where the information pushing system based on blockchain offline payment includes a cloud service pushing platform and a plurality of digital financial service terminals in communication connection with the cloud service pushing platform;

the cloud service pushing platform is used for receiving an information pushing request based on offline payment sent by the digital financial service terminal and obtaining user subscription label information of the information pushing request; wherein the user subscription tag information comprises subscription tag items;

the cloud service push platform is used for determining a business push object list corresponding to the subscription tag item according to big data mining information of the subscription tag item, wherein the big data mining information is obtained by carrying out big data mining on an offline bill data set generated by a digital financial service terminal in a block chain offline payment environment and a target payment environment element corresponding to the offline bill data set by the cloud service push platform;

and the cloud service pushing platform is used for pushing the business pushing directional content corresponding to the business pushing object list to the digital financial service terminal.

In a fourth aspect, an embodiment of the present application further provides a cloud service pushing platform, where the cloud service pushing platform includes a processor, a machine-readable storage medium, and a network interface, where the machine-readable storage medium, the network interface, and the processor are connected through a bus system, the network interface is used for being in communication connection with at least one digital financial service terminal, the machine-readable storage medium is used for storing a program, an instruction, or a code, and the processor is used for executing the program, the instruction, or the code in the machine-readable storage medium to execute the information pushing method based on block chain offline payment in the first aspect or any one of possible implementation manners in the first aspect.

In a fifth aspect, an embodiment of the present application provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instructions are executed, the computer executes the information pushing method based on blockchain offline payment in the first aspect or any one of the possible implementations of the first aspect.

Based on any one of the above aspects, the information pushing request based on offline payment sent by the digital financial service terminal is used for obtaining the user subscription tag information of the information pushing request, and then information is mined according to the big data of the subscription tag item to determine the service pushing object list corresponding to the subscription tag item, so that the oriented content is pushed to the service corresponding to the service pushing object list of the digital financial service terminal. Therefore, the oriented content can be effectively pushed to the business corresponding to the business pushing object list pushed by the digital financial service terminal according to the offline bill data condition of the offline payment scene, and the coverage of information pushing is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are required to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered as limiting the scope, and it will be apparent to those skilled in the art that other related drawings can be obtained from the drawings without inventive effort.

Fig. 1 is a schematic application scenario diagram of an information push system based on offline payment of a blockchain according to an embodiment of the present disclosure;

fig. 2 is a schematic flowchart of an information pushing method based on offline payment with blockchain according to an embodiment of the present disclosure;

fig. 3 is a functional module diagram of an information pushing apparatus based on offline payment of a blockchain according to an embodiment of the present disclosure;

fig. 4 is a schematic block diagram of structural components of a cloud service push platform for implementing the information push method based on blockchain offline payment according to the embodiment of the present application.

Detailed Description

The present application will now be described in detail with reference to the drawings, and the specific operations in the method embodiments may also be applied to the apparatus embodiments or the system embodiments.

Fig. 1 is an interaction diagram of aninformation push system 10 based on offline payment of a blockchain according to an embodiment of the present application. Theinformation push system 10 based on blockchain offline payment may include a cloudservice push platform 100 and a digitalfinancial service terminal 200 communicatively connected to the cloudservice push platform 100. Theinformation push system 10 based on blockchain offline payment shown in fig. 1 is only one possible example, and in other possible embodiments, theinformation push system 10 based on blockchain offline payment may also include only one of the components shown in fig. 1 or may also include other components.

In this embodiment, the digital financial services terminal 200 may comprise 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, an augmented reality device, or the like, or any combination thereof. In some embodiments, the internet of things device may include a control device of a smart appliance device, a smart monitoring device, a smart television, a smart camera, and the like, or any combination thereof. In some embodiments, the wearable device may include a smart bracelet, a smart lace, smart glass, a smart helmet, a smart watch, a smart garment, a smart backpack, a smart accessory, or the like, or any combination thereof. In some embodiments, the smart mobile device may include a smartphone, a personal digital assistant, a gaming device, and the like, or any combination thereof. In some embodiments, the virtual reality device and the augmented reality device may include a virtual reality helmet, virtual reality glass, a virtual reality patch, an augmented reality helmet, augmented reality glass, an augmented reality patch, 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.

In this embodiment, the cloudservice push platform 100 and the digitalfinancial service terminal 200 in theinformation push system 10 based on blockchain offline payment may cooperatively perform the information push method based on blockchain offline payment described in the following method embodiment, and for a specific step part of the cloudservice push platform 100 and the digitalfinancial service terminal 200, reference may be made to the detailed description of the following method embodiment.

Based on the inventive concept of the technical scheme provided by the application, the cloudservice push platform 100 provided by the application can be applied to scenes such as smart medical treatment, smart city management, smart industrial internet, general service monitoring management and the like, which can apply a big data technology or a cloud computing technology, and the like, and can also be applied to scenes such as but not limited to new energy automobile system management, smart cloud office, cloud platform data processing, cloud game data processing, cloud live broadcast processing, cloud automobile management platform, block chain financial data service platform and the like, but is not limited thereto.

In order to solve the technical problem in the foregoing background art, fig. 2 is a schematic flow chart of an information pushing method based on blockchain offline payment according to an embodiment of the present disclosure, where the information pushing method based on blockchain offline payment according to the present disclosure may be executed by the cloudservice pushing platform 100 shown in fig. 1, and the information pushing method based on blockchain offline payment is described in detail below.

Step S110, receiving an information push request based on offline payment sent by the digitalfinancial service terminal 200, and obtaining the user subscription tag information of the information push request.

And step S120, determining a service push object list corresponding to the subscription label item according to the big data mining information of the subscription label item.

Step S130, pushing the service push oriented content corresponding to the service push object list to the digitalfinancial service terminal 200.

In this embodiment, the user subscription tag information may include, for example, a subscription tag item, where the subscription tag item may be used to indicate a type of a subscribed service category, and may be specifically selected and configured in advance by a user of the digital financial services terminal 200, which is not limited in detail herein.

In this embodiment, the big data mining information may be big data mining information obtained by a cloud service push platform performing big data mining on an offline bill data set generated by the digitalfinancial service terminal 200 in a block chain offline payment environment and a target payment environment element corresponding to the offline bill data set. The offline billing data may refer to a scenario that, for each offline payment, a plurality of billing statistics boards are usually included, and may include, for example and without limitation, billing content, billing scenario, billing situation, and the like. The target payment environment element may be used to represent an environment element obtained offline in a specific payment process, such as a payment service scenario type, a user type of a payment user, and the like.

Based on the above steps, in this embodiment, the user subscription tag information of the information push request is obtained through the information push request based on offline payment sent by the digitalfinancial service terminal 200, and then the information is mined according to the big data of the subscription tag item, so as to determine the service push object list corresponding to the subscription tag item, thereby pushing the service push directional content corresponding to the service push object list to the digitalfinancial service terminal 200. Therefore, the oriented content of the business push corresponding to the business push object list can be effectively pushed to the digitalfinancial service terminal 200 according to the offline bill data condition of the offline payment scene, and the coverage of information push is improved.

In one possible implementation, step S120 may be implemented by the following exemplary substeps, which are described in detail below.

And a substep S121, obtaining the potential demand information and the demand service scene information of the potential demand information from the big data mining information of the subscription label item. The demand analysis templates used by the potential demand information and the demand service scenario information are both first demand analysis templates.

And a substep S122, processing the potential demand information according to the demand service scene information, and generating demand extension information of the potential demand information.

And a substep S123 of performing theme-oriented feature extraction on the potential demand information and the demand extension information, and determining second theme feature information corresponding to the first theme feature information corresponding to the demand extension information from the extracted current theme-oriented feature information.

And a substep S124 of performing feature fusion on the first theme feature information and the second theme feature information to obtain third theme feature information.

And a substep S125, outputting target topic distribution information corresponding to the potential demand information according to the third topic feature information.

For example, the requirement analysis template used by the target topic distribution information is a second requirement analysis template. Therefore, topic distribution conversion can be carried out based on different requirement analysis templates, so that different types of requirement analysis templates can be defined by a user during subsequent pushing.

Exemplarily, in the sub-step S122, it can be realized by the following specific embodiments.

(1) And performing theme-oriented feature extraction on the potential demand information, performing theme entity object identification on the first theme-oriented feature corresponding to the obtained potential demand information, and obtaining a first entity object set corresponding to the potential demand information according to the identified theme entity object.

(2) And performing theme-oriented feature extraction on the demand service scene information, performing theme entity object identification on a second theme-oriented feature corresponding to the acquired demand service scene information, and acquiring a second entity object set corresponding to the demand service scene information according to the identified theme entity object.

(3) Obtaining first entity relation information stored in the first entity object set, and converting the first entity relation information into a corresponding first entity relation vector.

(4) And obtaining second entity relationship information respectively stored by a plurality of entity key nodes in the second entity object set, and converting each second entity relationship information into a corresponding second entity relationship vector.

(5) And calculating the vector inner product result of each second entity relationship vector and the first entity relationship vector.

(6) And sorting the vector inner product result corresponding to each second entity relationship vector, and selecting a plurality of similar entity relationship vectors from the plurality of second entity relationship vectors according to the sorting result.

(7) And performing convolution processing on the plurality of similar entity relationship vectors to obtain convolution relationship vectors.

(8) And calculating the entity service coverage coincidence degree of the first entity object set and the second entity object set, and obtaining an influence factor vector according to the calculated entity service coverage coincidence degree. The influence factor vector includes a weight parameter corresponding to each entity key node in the second entity object set.

(9) And calculating the vector inner product of the convolution relation vector and the influence factor vector, and taking the calculated result as the expansion service vector of the first entity relation information.

(10) And adding the expanded service vector to a potential demand unit set in the potential demand information to obtain initial expanded data.

(11) And performing subject entity object identification on the initial extension data to obtain a reference subject entity object.

(12) And obtaining demand extension information corresponding to the potential demand information according to the first entity object set, the second entity object set and the reference subject entity object.

Exemplarily, in the sub-step S123, it can be realized by the following specific embodiments.

(1) And extracting theme-oriented features of the potential demand information and the demand extension information to obtain current theme-oriented feature information mapped in the theme-oriented features of the potential demand information and the demand extension information.

In this embodiment, the current theme-oriented feature information includes distribution feature information of a plurality of theme feature nodes.

(2) And determining similar distribution characteristic information of the first topic characteristic information from the distribution characteristic information of the plurality of topic characteristic nodes contained in the current topic-oriented characteristic information, and taking the similar distribution characteristic information as second topic characteristic information.

Exemplarily, in the sub-step S124, it can be realized by the following specific embodiments.

(1) And respectively inputting the first theme characteristic information and the second theme characteristic information into a preset theme distribution analysis model so that the theme distribution analysis model respectively outputs the respective effective theme characteristic information of the first theme characteristic information and the second theme characteristic information to obtain first target theme characteristic information and second target theme characteristic information.

(2) And performing convolution calculation on the first target theme characteristic information to obtain first theme convolution information. And performing theme-oriented feature extraction on the first target theme feature information, performing convolution calculation on the extracted theme feature information to obtain second theme convolution information, and calculating the dot product of the first theme convolution information and the second theme convolution information to obtain a first theme feature set corresponding to the first target theme feature information.

(3) And performing convolution calculation on the second target theme characteristic information to obtain third theme convolution information. And performing theme-oriented feature extraction on the second target theme feature information, performing convolution calculation on the extracted theme feature information to obtain fourth theme convolution information, and calculating the dot product of the third theme convolution information and the fourth theme convolution information to obtain a second theme feature set corresponding to the second target theme feature information.

(4) And calculating a fusion feature set of the first topic feature set and the second topic feature set, and taking the obtained fusion feature set as third topic feature information.

Exemplarily, in the sub-step S125, it can be realized by the following specific embodiments.

(1) And acquiring a plurality of topic distribution description vectors in the third topic feature information and a topic distribution analysis strategy corresponding to each topic distribution description vector, wherein the topic distribution description vectors comprise a first topic distribution description vector and a second topic distribution description vector, and the first topic distribution description vector and the second topic distribution description vector are mapping association topic distribution description vector pairs existing between each other.

(2) And coding the third topic characteristic information to output a first basic coding vector corresponding to each first topic distribution description vector and a target coding vector corresponding to the third topic characteristic information.

(3) And calculating the association degree between the target coding vector and each first basic coding vector to obtain the coding vector proximity degree between each corresponding first theme distribution description vector and the third theme characteristic information.

(4) All description elements in each first theme distribution description vector and the corresponding business description weight of each first theme distribution description vector are identified.

(5) And generating a theme distribution clustering node of the corresponding theme distribution description vector according to all the description elements and the service description weight.

(6) And generating a theme distribution clustering range corresponding to each theme distribution description vector according to the theme distribution analysis strategy.

As an example, the topic distribution parsing policy may include a vector structure clustering policy for topic distribution description vectors and a feature value clustering policy for topic distribution description vectors. Based on the theme distribution description vector clustering method, the theme distribution description vector can be subjected to characteristic value clustering according to the characteristic value clustering strategy to generate theme distribution information characteristic values corresponding to the theme distribution description vector, then the theme distribution description vector can be subjected to vector structure clustering according to the vector structure clustering strategy to generate vector structure clustering results corresponding to the theme distribution description vector, and therefore a theme distribution clustering range corresponding to the theme distribution description vector can be generated according to the theme distribution information characteristic values and the vector structure clustering results.

(7) And obtaining a respective first clustering group of each topic distribution description vector by using each topic distribution clustering node corresponding to the topic distribution clustering range.

(8) And clustering the first basic coding vector corresponding to each topic distribution description vector according to the second topic distribution description vector to obtain a second clustering group of each topic distribution description vector.

(9) And determining a target topic distribution description vector in the plurality of topic distribution description vectors according to the first clustering group and the second clustering group.

(10) And taking the first basic coding vector corresponding to the target topic distribution description vector as a second coding vector, and adding the second coding vector into the third topic characteristic information to output target topic distribution information corresponding to the potential demand information.

Exemplarily, the topic distribution description vector further includes a third topic distribution description vector, and before (7), the encoding translation parameter of each first basic encoding vector with respect to each reference encoding object in the reference encoding object list corresponding to the third topic distribution description vector may be further calculated, and then all the encoding translation parameters corresponding to each first basic encoding vector are fused to obtain a fused encoding parameter corresponding to each first basic encoding vector.

On this basis, all the first basic coding vectors may be arranged in sequence according to the fusion coding parameter corresponding to each first basic coding vector, the respective priority parameter of each first basic coding vector is determined according to the arranged sequence of each first basic coding vector, and then the fusion coding parameter corresponding to each first basic coding vector is processed according to the respective priority parameter of each first basic coding vector, so as to generate the weighted fusion coding parameter of each topic distribution description vector.

Thus, in (7), the weighted fusion coding parameters of each topic distribution description vector may be clustered by using each topic distribution clustering node corresponding to the topic distribution clustering range, so as to obtain a first clustering group corresponding to each topic distribution description vector.

In a possible implementation manner, before step S120, the information pushing method based on blockchain offline payment according to this embodiment may further include step S101 of obtaining big data mining information of the subscription tag item.

For example, step S101 may be specifically realized by the following substeps.

In sub-step S1011, an offline billing data set generated by each digitalfinancial service terminal 200 in the blockchain offline payment environment and a target payment environment element corresponding to the offline billing data set are obtained from each digitalfinancial service terminal 200.

The target payment environment element may be used to represent an environment element obtained offline in a specific payment process, such as a payment service scenario type, a user type of a payment user, and the like.

Step S1012, acquiring the mineable target service of the service tag to be mined under the target payment environment element, grouping the mineable target service under each target payment environment element according to a predetermined subscribed push group, and generating a mineable target service set of each subscribed push group respectively.

The offline billing data may refer to a scenario that, for each offline payment, a plurality of billing statistics boards are usually included, and may include, for example and without limitation, billing content, billing scenario, billing situation, and the like.

The mineable target service can be used for representing a mining type label of a specific application of the service label to be mined under each target payment environment element of the offline bill data, such as a fresh knowledge new mining type label, a digital product knowledge new mining type label and the like.

In this embodiment, the predetermined subscribed push group may be flexibly selected according to actual design requirements, and is mainly used to represent a subscription push selection menu provided for different users, which is not limited herein in detail.

Step S1013, for each subscribed push group, acquiring the knowledge graph data of each mineable target service in the mineable target service set of the subscribed push group, which matches the offline bill data set, and 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 to obtain big data mining information of each subscribed push group;

step S1014, obtaining the big data mining information of the target subscribed push group included in the subscription tag item from the big data mining information of each subscribed push group.

Based on the above steps, the mining-capable target service of the service tag to be mined under the target payment environment element corresponding to the offline bill data set is considered, and then the mining-capable target service under each target payment environment element is grouped based on the predetermined subscribed push group, so that differences between different target payment environment elements and the subscribed push groups are considered, and thus, the big data mining is performed on the knowledge graph data set of each subscribed push group based on the push service graph corresponding to the subscribed push group, the accuracy of the big data mining can be effectively improved, and the big data mining result can be more matched with an actual service scene.

In one possible implementation, such as for step S1013, in the process of obtaining the knowledge-graph data that each mineable target service in the set of mineable target services of the subscribed push group matches the offline billing data set, the following exemplary sub-steps may be further implemented, which are described in detail below.

And a substep S10131, obtaining a matching keyword vector related to each mineable target service in the mineable target service set subscribed to the push packet.

And a substep S10132 of matching the corresponding bill plate contents from the offline bill data set according to the matching keyword vector associated with each mineable target service.

And a substep S10133 of determining, according to the knowledge-graph content corresponding to each service record plate in the bill plate content matched with the matching keyword vector related to each mineable target service, that each mineable target service in the mineable target service set of the subscribed push packet matches the knowledge-graph data of the offline bill data set.

In one possible implementation, for example, for step S1013, in the process of performing big data mining on the knowledge-graph data set of each subscribed push group based on the push service graph corresponding to the subscribed push group, the following sub-steps may be performed.

In step S10134, 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 are determined based on the push service portrait corresponding to the subscribed push group.

In the substep S10135, mining flow parameters of the big data mining component required for big data mining of the push service portrait node in each subscribed push group are determined 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.

In the sub-step S10136, according to the mining flow parameter of the big data mining component required by each push service portrait node, each big data mining component is determined as a mining unit, and the operation configuration information corresponding to the mining unit is operation configuration information included in the push service portrait node and other than the currently configured operation configuration information of the subscribed push packet.

And a substep S10137 of establishing a mining business relationship of the mining unit according to the running configuration information corresponding to the mining unit, determining a business matching element of the mining business relationship, and obtaining preliminary mining information of the first mining unit in the business matching element for performing big data mining on each subscribed pushed and grouped knowledge graph data set.

And a substep S10138 of screening the primary mining information of the mining unit and the mining units behind the mining unit when screening the primary mining information of each mining unit behind the first mining unit in sequence according to the hierarchy of the mining unit, reestablishing the mining business relationship of the mining unit according to the screened primary mining information, determining the business matching elements of the reestablished mining business relationship, and obtaining the screened primary mining information of the mining unit in the business matching elements of the reestablished mining business relationship.

And a substep S10139 of obtaining the screened preliminary mining information of all the mining units, and taking the screened preliminary mining information of all the mining units as a big data mining result.

In one possible implementation, for example, after step S1013, the following steps may be further included:

step S1015, judging whether the extended loading subject feature information for indicating that the extensible loading service exists in the extensible target service exists in the big data mining process, and extracting a first knowledge graph of a first extensible target service corresponding to the extended loading subject feature information of the big data mining and at least one second knowledge graph of a second extensible target service having an extended loading service relation with the first extensible target service when the extended loading subject feature information is detected;

and step S1016, determining global big data mining information between the first knowledge graph and the at least one second knowledge graph according to a preset artificial intelligence model.

In one possible implementation, for example, for step S1015, a first knowledge graph of a first mineable target service corresponding to the extended loading topic feature information of the big data mining and a second knowledge graph of at least one second mineable target service having an extended loading business relationship with the first mineable target service may be extracted from the big data mining record information generated in the big data mining process. Wherein the at least one second mineable target service for which an extended load business relationship exists with the first mineable target service may refer to a second mineable target service for which a linkage effect associated with the first mineable target service exists.

For example, if a certain mineable target service needs to extend mining during the mining of a first mineable target service, then the mineable target service may be understood as a second mineable target service that has an extended load business relationship with the first mineable target service.

In one possible implementation, such as for step S1016, this may be achieved by the following exemplary substeps, described in detail below.

And a substep S10161 of fusing the knowledge graph nodes corresponding to the first knowledge graph and the at least one second knowledge graph according to each identical knowledge graph node to obtain a fused knowledge graph.

The sub-step S10162 adds the first knowledge-map and the at least one second knowledge-map to a preset data map classification queue, and establishes a plurality of first data map classification parameters of the first knowledge-map and a plurality of second data map classification parameters of the second knowledge-map based on the data map classification queue.

Substep S10163, determining first knowledge expression information of a first excavatable target service according to each first data map classification parameter, determining second knowledge expression information of a second excavatable target service according to each second data map classification parameter, mapping the first knowledge expression information and the second knowledge expression information to a knowledge entity feature model to obtain a first knowledge map feature corresponding to the first knowledge expression information and a second knowledge map feature corresponding to the second knowledge expression information, determining a plurality of knowledge corpus objects corresponding to a fusion knowledge map of the knowledge entity feature model, summarizing the plurality of knowledge corpus objects to obtain at least a plurality of different classes of knowledge corpus excavation lists, excavating first corpus characterization features corresponding to the first knowledge map feature and corresponding second knowledge map characterization features of each knowledge corpus object in the knowledge corpus excavation list in a preset big data excavation process aiming at each knowledge corpus excavation list The second corpus of features depicts features.

And a substep S10164 of merging the generated simulated mining streams according to the mining results of the first corpus portrait characterization feature and the second corpus portrait characterization feature corresponding to each knowledge corpus object in the knowledge corpus mining list and the preset priority of the knowledge expectation, restoring the merged simulated mining streams, and determining global big data mining information of the first excavatable target service and the at least one second excavatable target service.

In this way, subsequent big data mining can be performed with the associated mineable target service as an independent mining target in a targeted manner in the actual big data mining process.

In one possible implementation manner, for step S10164, for example, in the process of restoring the spliced simulated mining stream and determining the global big data mining information of the first mineable target service and the at least one second mineable target service, the spliced simulated mining stream may be reversely converted according to each corresponding simulated mining node to obtain the global big data mining information of the first mineable target service and the at least one second mineable target service.

Fig. 3 is a schematic diagram of functional modules of theinformation pushing apparatus 300 based on blockchain offline payment according to an embodiment of the present disclosure, in this embodiment, theinformation pushing apparatus 300 based on blockchain offline payment may be divided into the functional modules according to an embodiment of a method performed by the cloudservice pushing platform 100, that is, the following functional modules corresponding to theinformation pushing apparatus 300 based on blockchain offline payment may be used to perform the method embodiments performed by the cloudservice pushing platform 100. Thedevice 300 for pushing information based on blockchain offline payment may include areceiving module 310, a determiningmodule 320, and a pushingmodule 330, and the functions of the functional modules of thedevice 300 for pushing information based on blockchain offline payment are described in detail below.

A receivingmodule 310, configured to receive an information push request based on offline payment sent by the digitalfinancial service terminal 200, and obtain user subscription tag information of the information push request; wherein the user subscription tag information comprises subscription tag items. The receivingmodule 310 may be configured to perform the step S110, and the detailed implementation of the receivingmodule 310 may refer to the detailed description of the step S110.

The determiningmodule 320 is configured to determine, according to big data mining information of the subscription tag item, a service push object list corresponding to the subscription tag item, where the big data mining information is obtained by the cloud service push platform performing big data mining on an offline bill data set generated by the digitalfinancial service terminal 200 in a block chain offline payment environment and a target payment environment element corresponding to the offline bill data set. The determiningmodule 320 may be configured to perform the step S120, and the detailed implementation of the determiningmodule 320 may refer to the detailed description of the step S120.

The pushingmodule 330 is configured to push the service push targeted content corresponding to the service push object list to the digitalfinancial service terminal 200. The pushingmodule 330 may be configured to perform the step S130, and the detailed implementation manner of the pushingmodule 330 may refer to the detailed description of the step S130.

It should be noted that the division of the modules of the above apparatus is only a logical division, and the actual implementation may be wholly or partially integrated into one physical entity, or may be physically separated. And these modules may all be implemented in software invoked by a processing element. Or may be implemented entirely in hardware. And part of the modules can be realized in the form of calling software by the processing element, and part of the modules can be realized in the form of hardware. For example, the receivingmodule 310 may be a processing element separately set up, or may be implemented by being integrated into a chip of the apparatus, or may be stored in a memory of the apparatus in the form of program code, and the processing element of the apparatus calls and executes the functions of the receivingmodule 310. Other modules are implemented similarly. In addition, all or part of the modules can be integrated together or can be independently realized. The processing element described herein may be an integrated circuit having signal processing capabilities. In implementation, each step of the above method or each module above may be implemented by an integrated logic circuit of hardware in a processor element or an instruction in the form of software.

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 (ASICs), or one or more microprocessors (DSPs), or one or more Field Programmable Gate Arrays (FPGAs), among others. For another example, when some of the above modules are implemented in the form of a processing element scheduler code, the processing element may be a general-purpose processor, such as a Central Processing Unit (CPU) or other processor that can call program code. As another example, these modules may be integrated together, implemented in the form of a system-on-a-chip (SOC).

Fig. 4 illustrates a hardware structure diagram of the cloudservice push platform 100 for implementing the control device, where the cloudservice push platform 100 provided by the embodiment of the present disclosure may include aprocessor 110, a machine-readable storage medium 120, abus 130, and atransceiver 140, as shown in fig. 4.

In an implementation process, at least oneprocessor 110 executes computer-executable instructions stored in the machine-readable storage medium 120 (for example, the receivingmodule 310, the determiningmodule 320, and the pushingmodule 330 included in theinformation pushing apparatus 300 based on blockchain offline payment shown in fig. 3), so that theprocessor 110 may execute the information pushing method based on blockchain offline payment according to the above method embodiment, where theprocessor 110, the machine-readable storage medium 120, and thetransceiver 140 are connected through thebus 130, and theprocessor 110 may be configured to control the transceiving action of thetransceiver 140, so as to perform data transceiving with the aforementioned digitalfinancial service terminal 200.

For a specific implementation process of theprocessor 110, reference may be made to the above-mentioned method embodiments executed by the cloudservice push platform 100, and implementation principles and technical effects thereof are similar, and details of this embodiment are not described herein again.

In the embodiment shown in fig. 4, it should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.

The machine-readable storage medium 120 may comprise high-speed RAM memory and may also include non-volatile storage NVM, such as at least one disk memory.

Thebus 130 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. Thebus 130 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.

In addition, the embodiment of the disclosure also provides a readable storage medium, in which computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the information push method based on the blockchain offline payment is implemented.

The readable storage medium described above may be implemented by any type of volatile or non-volatile memory 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 or optical disk. Readable storage media can be any available media that can be accessed by a general purpose or special purpose computer.

Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the present disclosure, and not for limiting the same; while the present disclosure has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present disclosure.

Claims (10)

Translated fromChinese

1.一种基于区块链离线支付的信息推送方法，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. an information push 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:接收所述数字金融服务终端发送的基于离线支付的信息推送请求，获得所述信息推送请求的用户订阅标签信息；其中，所述用户订阅标签信息包括订阅标签项目；Receive an information push request based on offline payment sent by the digital financial service terminal, and obtain user subscription tag information of the information push request; wherein, the user subscription tag information includes subscription tag items;根据所述订阅标签项目的大数据挖掘信息，确定出所述订阅标签项目对应的业务推送对象列表，其中，所述大数据挖掘信息为所述云服务推送平台对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进行大数据挖掘获得的大数据挖掘信息；According to the big data mining information of the subscription tag item, the list of business push objects corresponding to the subscription tag item is determined, wherein the big data mining information is the cloud service push platform for the digital financial service terminal in the blockchain. The offline bill data set generated in the offline payment environment and the big data mining information obtained by performing big data mining on the target payment environment element corresponding to the offline bill data set;向所述数字金融服务终端推送所述业务推送对象列表对应的业务推送定向内容；Pushing the business push targeted content corresponding to the business push object list to the digital financial service terminal;其中，所述云服务推送平台在获取所述订阅标签项目的大数据挖掘信息之前，从所述数字金融服务终端中获取所述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 obtains, from the digital financial service terminal, the offline bill data generated by the digital financial service terminal in the blockchain offline payment environment before obtaining the big data mining information of the subscription tag item. collection and the target payment environment element corresponding to the offline bill data collection.2.根据权利要求1所述的基于区块链离线支付的信息推送方法，其特征在于，所述根据所述订阅标签项目的大数据挖掘信息，确定出所述订阅标签项目对应的业务推送对象列表的步骤，包括：2. The method for pushing information based on blockchain offline payment according to claim 1, characterized in that, according to the big data mining information of the subscription tag item, the service push object corresponding to the subscription tag item is determined. List of steps, including:从所述订阅标签项目的大数据挖掘信息中获取潜在需求信息和所述潜在需求信息的需求业务场景信息；所述潜在需求信息和所述需求业务场景信息使用的需求分析模板均为第一需求分析模板；Obtain potential demand information and demand business scenario information of the potential demand information from the big data mining information of the subscription tag item; the potential demand information and the demand analysis template used by the demand business scene information are both the first demand Analysis template;根据所述需求业务场景信息对所述潜在需求信息进行处理，生成所述潜在需求信息的需求扩展信息；Process the potential demand information according to the demand business scenario information, and generate demand expansion information of the potential demand information;对所述潜在需求信息和所述需求扩展信息进行面向主题特征提取，从提取得到的当前面向主题特征信息中确定与所述需求扩展信息对应的第一主题特征信息对应的第二主题特征信息；Subject-oriented feature extraction is performed on the potential demand information and the demand expansion information, and second theme feature information corresponding to the first theme feature information corresponding to the demand expansion information is determined from the extracted current theme-oriented feature information;对所述第一主题特征信息和所述第二主题特征信息进行特征融合，得到第三主题特征信息；Performing feature fusion on the first theme feature information and the second theme feature information to obtain third theme feature information;根据所述第三主题特征信息输出所述潜在需求信息对应的目标主题分布信息；所述目标主题分布信息使用的需求分析模板为第二需求分析模板。The target topic distribution information corresponding to the potential demand information is output according to the third topic feature information; the demand analysis template used by the target topic distribution information is the second demand analysis template.3.根据权利要求2所述的基于区块链离线支付的信息推送方法，其特征在于，所述根据所述需求业务场景信息对所述潜在需求信息进行处理，生成所述潜在需求信息的需求扩展信息的步骤，包括：3. The method for information pushing based on blockchain offline payment according to claim 2, wherein the potential demand information is processed according to the demand business scenario information to generate the demand for the potential demand information Steps to expand the information, including:对所述潜在需求信息进行面向主题特征提取，对得到的所述潜在需求信息对应的第一面向主题特征进行主题实体对象识别，根据所识别到的主题实体对象得到所述潜在需求信息对应的第一实体对象集合；Subject-oriented feature extraction is performed on the potential demand information, subject entity object recognition is performed on the first subject-oriented feature corresponding to the obtained potential demand information, and the first subject entity object corresponding to the potential demand information is obtained according to the identified subject entity object. a collection of entity objects;对所述需求业务场景信息进行面向主题特征提取，对得到的所述需求业务场景信息对应的第二面向主题特征进行主题实体对象识别，根据所识别到的主题实体对象得到所述需求业务场景信息对应的第二实体对象集合；Subject-oriented feature extraction is performed on the demand business scene information, subject entity object recognition is performed on the second subject-oriented feature corresponding to the obtained demand business scene information, and the demand business scene information is obtained according to the identified subject entity object. the corresponding second entity object set;获取所述第一实体对象集合中保存的第一实体关系信息，以及将所述第一实体关系信息转换为对应的第一实体关系向量；Obtaining the first entity relationship information stored in the first entity object set, and converting the first entity relationship information into a corresponding first entity relationship vector;获取所述第二实体对象集合中多个实体关键节点各自保存的第二实体关系信息，以及将每个所述第二实体关系信息转换为对应的第二实体关系向量；Acquiring second entity relationship information stored by multiple entity key nodes in the second entity object set, and converting each of the second entity relationship information into a corresponding second entity relationship vector;计算每个所述第二实体关系向量与所述第一实体关系向量的向量内积结果；Calculate the vector inner product result of each of the second entity relationship vectors and the first entity relationship vectors;对每个所述第二实体关系向量对应的向量内积结果进行排序，根据排序结果从多个所述第二实体关系向量中选出多个相似实体关系向量；Sorting the vector inner product results corresponding to each of the second entity relationship vectors, and selecting a plurality of similar entity relationship vectors from a plurality of the second entity relationship vectors according to the sorting results;对所述多个相似实体关系向量进行卷积处理，得到卷积关系向量；Perform convolution processing on the plurality of similar entity relationship vectors to obtain convolution relationship vectors;对所述第一实体对象集合和所述第二实体对象集合进行实体业务覆盖重合度计算，根据计算得到的实体业务覆盖重合度得到影响因子向量；所述影响因子向量中包含有所述第二实体对象集合中各个实体关键节点对应的权重参数；Performing entity business coverage coincidence degree calculation on the first entity object set and the second entity object set, and obtaining an impact factor vector according to the calculated entity business coverage coincidence degree; the impact factor vector includes the second Weight parameters corresponding to each entity key node in the entity object set;计算所述卷积关系向量和所述影响因子向量两者的向量内积，并将计算出的结果作为所述第一实体关系信息的扩展业务向量；Calculate the vector inner product of the convolution relationship vector and the impact factor vector, and use the calculated result as the extended service vector of the first entity relationship information;将所述扩展业务向量添加至所述潜在需求信息中设定的潜在需求单元，得到初始扩展数据；adding the expanded service vector to the potential demand unit set in the potential demand information to obtain initial expanded data;对所述初始扩展数据进行主题实体对象识别，得到参考主题实体对象；Perform subject entity object identification on the initial extended data to obtain a reference subject entity object;根据所述第一实体对象集合、所述第二实体对象集合以及所述参考主题实体对象，得到所述潜在需求信息对应的所述需求扩展信息。According to the first entity object set, the second entity object set and the reference subject entity object, the demand extension information corresponding to the potential demand information is obtained.4.根据权利要求2所述的基于区块链离线支付的信息推送方法，其特征在于，所述对所述潜在需求信息和所述需求扩展信息进行面向主题特征提取，从提取得到的当前面向主题特征信息中确定与所述需求扩展信息对应的第一主题特征信息对应的第二主题特征信息的步骤，包括：4. The method for information pushing based on blockchain offline payment according to claim 2, wherein the subject-oriented feature extraction is performed on the potential demand information and the demand expansion information, and the extracted current The step of determining the second theme feature information corresponding to the first theme feature information corresponding to the demand extension information in the theme feature information includes:对所述潜在需求信息和所述需求扩展信息进行面向主题特征提取，得到所述潜在需求信息和所述需求扩展信息的面向主题特征中映射的当前面向主题特征信息；所述当前面向主题特征信息包括多个主题特征节点的分布特征信息；Subject-oriented feature extraction is performed on the potential demand information and the demand extension information to obtain current theme-oriented feature information mapped in the topic-oriented features of the potential demand information and the demand expansion information; the current theme-oriented feature information Including distribution feature information of multiple topic feature nodes;从所述当前面向主题特征信息包含的多个主题特征节点的分布特征信息中确定所述第一主题特征信息的相似分布特征信息，并将所述相似分布特征信息作为所述第二主题特征信息。Determine the similar distribution feature information of the first theme feature information from the distribution feature information of a plurality of theme feature nodes included in the current theme-oriented feature information, and use the similar distribution feature information as the second theme feature information .5.根据权利要求2所述的基于区块链离线支付的信息推送方法，其特征在于，所述对所述第一主题特征信息和所述第二主题特征信息进行特征融合，得到第三主题特征信息的步骤，包括：5. The method for information pushing based on blockchain offline payment according to claim 2, wherein the feature fusion of the first theme feature information and the second theme feature information is performed to obtain a third theme Steps to characterize information, including:将所述第一主题特征信息和所述第二主题特征信息分别输入预设的主题分布解析模型中，以使所述主题分布解析模型分别输出所述第一主题特征信息和所述第二主题特征信息各自的有效主题特征信息，得到第一目标主题特征信息和第二目标主题特征信息；Inputting the first topic feature information and the second topic feature information into a preset topic distribution analysis model respectively, so that the topic distribution analysis model outputs the first topic feature information and the second topic respectively The respective valid theme feature information of the feature information, to obtain the first target theme feature information and the second target theme feature information;对所述第一目标主题特征信息进行卷积计算，得到第一主题卷积信息；对所述第一目标主题特征信息进行面向主题特征提取，并对提取得到的主题特征信息进行卷积计算，得到第二主题卷积信息，计算所述第一主题卷积信息和所述第二主题卷积信息的点乘，得到与所述第一目标主题特征信息对应的第一主题特征集；Perform convolution calculation on the first target topic feature information to obtain first topic convolution information; perform topic-oriented feature extraction on the first target topic feature information, and perform convolution calculation on the extracted topic feature information, Obtain the second topic convolution information, calculate the dot product of the first topic convolution information and the second topic convolution information, and obtain the first topic feature set corresponding to the first target topic feature information;对所述第二目标主题特征信息进行卷积计算，得到第三主题卷积信息；对所述第二目标主题特征信息进行面向主题特征提取，并对提取得到的主题特征信息进行卷积计算，得到第四主题卷积信息，计算所述第三主题卷积信息和所述第四主题卷积信息的点乘，得到与所述第二目标主题特征信息对应的第二主题特征集；Perform convolution calculation on the second target topic feature information to obtain third topic convolution information; perform topic-oriented feature extraction on the second target topic feature information, and perform convolution calculation on the extracted topic feature information, Obtain the fourth topic convolution information, calculate the dot product of the third topic convolution information and the fourth topic convolution information, and obtain the second topic feature set corresponding to the second target topic feature information;计算所述第一主题特征集和所述第二主题特征集两者的融合特征集，将得到的融合特征集作为所述第三主题特征信息。A fusion feature set of both the first theme feature set and the second theme feature set is calculated, and the obtained fusion feature set is used as the third theme feature information.6.根据权利要求2所述的基于区块链离线支付的信息推送方法，其特征在于，所述根据所述第三主题特征信息输出所述潜在需求信息对应的目标主题分布信息的步骤，包括：6. The method for information pushing based on blockchain offline payment according to claim 2, wherein the step of outputting the target topic distribution information corresponding to the potential demand information according to the third topic feature information comprises: :获取所述第三主题特征信息中的多条主题分布描述向量、以及与每个主题分布描述向量对应的主题分布解析策略，所述主题分布描述向量包括第一主题分布描述向量和第二主题分布描述向量，其中，所述第一主题分布描述向量和第二主题分布描述向量为相互之间存在映射关联主题分布描述向量对；Obtain a plurality of topic distribution description vectors in the third topic feature information and a topic distribution analysis strategy corresponding to each topic distribution description vector, where the topic distribution description vector includes a first topic distribution description vector and a second topic distribution a description vector, wherein the first topic distribution description vector and the second topic distribution description vector are a pair of subject distribution description vectors that are mapped to each other;对所述第三主题特征信息进行编码，以输出每个第一主题分布描述向量对应的第一基础编码向量、以及所述第三主题特征信息对应的目标编码向量；Encoding the third topic feature information to output the first basic encoding vector corresponding to each first topic distribution description vector and the target encoding vector corresponding to the third topic feature information;计算所述目标编码向量和每个所述第一基础编码向量之间的关联度，以得到对应的每条第一主题分布描述向量和所述第三主题特征信息之间的编码向量接近程度；Calculate the degree of association between the target encoding vector and each of the first basic encoding vectors, to obtain the degree of proximity of the encoding vectors between the corresponding first topic distribution description vectors and the third topic feature information;识别出每个第一主题分布描述向量中的所有描述元素、以及每个第一主题分布描述向量各自对应的业务描述权重；Identifying all description elements in each first topic distribution description vector and the business description weight corresponding to each first topic distribution description vector;根据所述所有描述元素和所述业务描述权重，生成对应的主题分布描述向量的主题分布聚类节点；According to all the description elements and the business description weight, generate a topic distribution clustering node of the corresponding topic distribution description vector;根据所述主题分布解析策略生成每个主题分布描述向量对应的主题分布聚类范围；Generate a topic distribution clustering range corresponding to each topic distribution description vector according to the topic distribution analysis strategy;利用所述主题分布聚类范围对应的每条主题分布聚类节点，得到每个主题分布描述向量各自的第一聚类分组；Use each topic distribution clustering node corresponding to the topic distribution clustering range to obtain the respective first cluster grouping of each topic distribution description vector;根据所述第二主题分布描述向量对每个主题分布描述向量对应的第一基础编码向量进行聚类，得到每个主题分布描述向量的第二聚类分组；According to the second topic distribution description vector, the first basic coding vector corresponding to each topic distribution description vector is clustered to obtain the second clustering grouping of each topic distribution description vector;根据所述第一聚类分组和第二聚类分组，在所述多条主题分布描述向量中确定出目标主题分布描述向量；According to the first cluster grouping and the second cluster grouping, a target topic distribution description vector is determined from the plurality of topic distribution description vectors;将所述目标主题分布描述向量对应的第一基础编码向量作为第二编码向量，并将所述第二编码向量添加至所述第三主题特征信息中，以输出所述潜在需求信息对应的目标主题分布信息。Taking the first basic encoding vector corresponding to the target topic distribution description vector as the second encoding vector, and adding the second encoding vector to the third topic feature information to output the target corresponding to the potential demand information Topic distribution information.7.根据权利要求6所述的基于区块链离线支付的信息推送方法，其特征在于，所述主题分布解析策略包括对所述主题分布描述向量的向量结构聚类策略、以及所述主题分布描述向量对应的特征值聚类策略；7. The method for information pushing based on blockchain offline payment according to claim 6, wherein the topic distribution analysis strategy comprises a vector structure clustering strategy describing vectors for the topic distribution, and the topic distribution Describe the eigenvalue clustering strategy corresponding to the vector;所述根据所述主题分布解析策略生成每个主题分布描述向量对应的主题分布聚类范围的步骤，包括：The step of generating the subject distribution clustering range corresponding to each subject distribution description vector according to the subject distribution analysis strategy includes:根据所述特征值聚类策略对所述主题分布描述向量进行特征值聚类，生成所述主题分布描述向量对应的主题分布信息特征值；Perform eigenvalue clustering on the subject distribution description vector according to the eigenvalue clustering strategy, and generate the subject distribution information eigenvalue corresponding to the subject distribution description vector;根据所述向量结构聚类策略对所述主题分布描述向量进行向量结构聚类，生成所述主题分布描述向量对应的向量结构聚类结果；Perform vector structure clustering on the subject distribution description vector according to the vector structure clustering strategy, and generate a vector structure clustering result corresponding to the subject distribution description vector;根据所述主题分布信息特征值和所述向量结构聚类结果，生成对应所述主题分布描述向量的主题分布聚类范围。According to the feature value of the topic distribution information and the vector structure clustering result, a topic distribution clustering range corresponding to the topic distribution description vector is generated.8.根据权利要求6所述的基于区块链离线支付的信息推送方法，其特征在于，所述主题分布描述向量还包括第三主题分布描述向量；8. The information push method based on blockchain offline payment according to claim 6, wherein the subject distribution description vector further comprises a third subject distribution description vector;在利用所述主题分布聚类范围对应的每个主题分布聚类节点，得到每个主题分布描述向量各自的第一聚类分组的步骤之前，所述方法还包括：Before using each topic distribution clustering node corresponding to the topic distribution clustering range to obtain the respective first cluster grouping of each topic distribution description vector, the method further includes:计算每个所述第一基础编码向量相对于所述第三主题分布描述向量所对应的参照编码对象列表中每个参照编码对象的编码平移参数；calculating the encoding translation parameter of each reference encoding object in the reference encoding object list corresponding to each of the first basic encoding vector relative to the third topic distribution description vector;对每个所述第一基础编码向量对应的所有编码平移参数进行融合，得到每个所述第一基础编码向量对应的融合编码参数；Fusing all coding translation parameters corresponding to each of the first basic coding vectors to obtain fused coding parameters corresponding to each of the first basic coding vectors;根据每个所述第一基础编码向量对应的融合编码参数，对所有所述第一基础编码向量按序排列，以根据每个所述第一基础编码向量排列后的顺序，确定每个所述第一基础编码向量各自的优先级参数；According to the fusion coding parameter corresponding to each of the first basic coding vectors, all the first basic coding vectors are arranged in order, so as to determine each of the first basic coding vectors according to the order in which the first basic coding vectors are arranged. respective priority parameters of the first basic coding vector;根据每个所述第一基础编码向量各自的优先级参数，对每个所述第一基础编码向量各自对应的融合编码参数进行处理，生成每个主题分布描述向量的加权融合编码参数；According to the respective priority parameters of each of the first basic coding vectors, the corresponding fusion coding parameters of each of the first basic coding vectors are processed to generate weighted fusion coding parameters of each topic distribution description vector;所述利用所述主题分布聚类范围对应的每个主题分布聚类节点，得到每个主题分布描述向量各自的第一聚类分组的步骤，包括：The step of using each topic distribution clustering node corresponding to the topic distribution clustering range to obtain the respective first cluster groupings of each topic distribution description vector includes:利用所述主题分布聚类范围对应的每个主题分布聚类节点对所述每个主题分布描述向量的加权融合编码参数进行聚类，得到每个主题分布描述向量对应的第一聚类分组。The weighted fusion coding parameters of each topic distribution description vector are clustered by using each topic distribution clustering node corresponding to the topic distribution clustering range to obtain a first clustering group corresponding to each topic distribution description vector.9.根据权利要求1-8中任意一项所述的基于区块链离线支付的信息推送方法，其特征在于，所述方法还包括：9. The method for information pushing based on blockchain offline payment according to any one of claims 1-8, wherein the method further comprises:获取所述订阅标签项目的大数据挖掘信息；obtaining the big data mining information of the subscription tag item;所述获取所述订阅标签项目的大数据挖掘信息的步骤，包括：The step of obtaining the big data mining information of the subscription tag item includes:从所述数字金融服务终端中获取所述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 the offline bill data set generated by the digital financial service terminal in the blockchain offline payment environment and the target payment environment element corresponding to the offline bill data set from the digital financial service terminal;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，得到各个已订阅推送分组的大数据挖掘信息；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, and obtains the big data mining information of each subscribed push group;从所述各个已订阅推送分组的大数据挖掘信息中获得所述订阅标签项目包括的目标已订阅推送分组的大数据挖掘信息。The big data mining information of the target subscribed push group included in the subscription tag item is obtained from the big data mining information of each subscribed push group.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信息推送方法。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. are connected through a bus system, the network interface is used for communicating with at least one digital financial service terminal, the machine-readable storage medium is used for storing programs, instructions or codes, and the processor is used for executing the machine-readable storage medium A program, instruction or code in the medium to execute the information push method based on blockchain offline payment of any one of claims 1-9.

Images