CN114625964A - Account recommendation method and device, computer equipment and storage medium - Google Patents
Account recommendation method and device, computer equipment and storage mediumDownload PDFInfo
- Publication number
- CN114625964A CN114625964ACN202210247532.XACN202210247532ACN114625964ACN 114625964 ACN114625964 ACN 114625964ACN 202210247532 ACN202210247532 ACN 202210247532ACN 114625964 ACN114625964 ACN 114625964A
- Authority
- CN
- China
- Prior art keywords
- account
- preset
- feature description
- information
- knowledge graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/3332—Query translation
- G06F16/3338—Query expansion
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
- G06F16/353—Clustering; Classification into predefined classes
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Artificial Intelligence (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Translated fromChinese
Description
Translated fromChinese技术领域technical field
本公开涉及数据处理技术领域，特别是涉及一种账户推荐的方法、装置、计算机设备和存储介质。The present disclosure relates to the technical field of data processing, and in particular, to a method, apparatus, computer device and storage medium for account recommendation.
背景技术Background technique
现有的个性化推荐技术中，在给用户进行账户推荐时，会根据用户的个性化偏好和需求筛选出相关的账户推荐给用户，这种方法通常是根据用户的历史行为数据进行账户的筛选和推荐。In the existing personalized recommendation technology, when recommending accounts to users, relevant accounts will be screened and recommended to users according to the user's personalized preferences and needs. This method is usually based on the user's historical behavior data to screen accounts. and recommended.
然而，根据用户的历史行为进行推荐会导致推荐账户重复缺乏新意等问题，并且在用户的历史行为数据较少的情况下，无法给用户进行准确的个性化账户推荐。However, recommending based on the user's historical behavior will lead to problems such as repetition of recommended accounts and lack of novelty, and when the user's historical behavior data is small, it is impossible to make accurate personalized account recommendations for the user.
发明内容SUMMARY OF THE INVENTION
基于此，有必要针对上述技术问题，提供一种优化推荐内容的账户推荐的方法、装置、计算机设备和存储介质。Based on this, it is necessary to provide a method, an apparatus, a computer device and a storage medium for optimizing the account recommendation of the recommended content in view of the above technical problems.
第一方面，本公开实施例提供了一种账户推荐的方法。所述方法包括：In a first aspect, an embodiment of the present disclosure provides a method for account recommendation. The method includes:
获取第一账户的特征信息，所述特征信息包括所述第一账户在注册时填写的兴趣信息以及所述第一账户的评论信息；Obtain characteristic information of the first account, where the characteristic information includes interest information filled in by the first account during registration and comment information of the first account;
根据所述特征信息，确定第二账户，其中，所述第二账户的特征信息与所述第一账户的特征信息的相似度在预设范围内；determining a second account according to the feature information, wherein the similarity between the feature information of the second account and the feature information of the first account is within a preset range;
从预设的账户知识图谱中确定与所述第二账户的特征描述语句相匹配的目标账户，将所述目标账户推荐至所述第一账户，其中，所述账户知识图谱包括账户与特征描述语句之间的关联关系。A target account matching the feature description sentence of the second account is determined from a preset account knowledge graph, and the target account is recommended to the first account, wherein the account knowledge graph includes account and feature descriptions relationship between sentences.
在其中一个实施例中，所述账户知识图谱的确定方法，包括：In one embodiment, the method for determining the account knowledge graph includes:
获取账户分类体系及特征描述语句，所述分类体系为根据账户的行业数据按照预设的分类规则处理得到；Obtaining an account classification system and a feature description statement, where the classification system is obtained by processing the industry data of the account according to preset classification rules;
根据所述分类体系及所述特征描述语句之间的关联关系，生成初级账户知识图谱；Generate a primary account knowledge graph according to the classification system and the relationship between the feature description sentences;
获取多个预设账户及所述多个预设账户的特征信息；acquiring multiple preset accounts and feature information of the multiple preset accounts;
根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，形成账户知识图谱。The relationship between the multiple preset accounts and the feature description sentence is determined according to the feature information, and the multiple preset accounts are added to the primary account knowledge graph to form an account knowledge graph.
在其中一个实施例中，所述特征描述语句的确定方式，包括：In one of the embodiments, the determination method of the feature description statement includes:
从所述行业数据中获取多个候选短语；obtain a plurality of candidate phrases from the industry data;
利用判别模型从所述多个候选短语确定特征描述语句，其中，所述判别模型为根据候选短语和候选短语对应的详细说明信息训练得到。The feature description sentences are determined from the plurality of candidate phrases by using a discriminant model, wherein the discriminant model is obtained by training according to the candidate phrases and the detailed description information corresponding to the candidate phrases.
在其中一个实施例中，所述从所述行业数据中获取多个候选短语，还包括：In one embodiment, the obtaining a plurality of candidate phrases from the industry data further includes:
对所述账户的行业数据进行分词处理，得到词语；Perform word segmentation processing on the industry data of the account to obtain words;
根据所述词语的语义按照预设的组合规则对所述词语进行组合，得到多个候选短语。According to the semantics of the words, the words are combined according to a preset combination rule to obtain a plurality of candidate phrases.
在其中一个实施例中，在所述根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，之后还包括：In one embodiment, after determining the relationship between the multiple preset accounts and the feature description sentence according to the feature information, adding the multiple preset accounts to the primary account knowledge graph, and then adding the multiple preset accounts to the primary account knowledge graph. Also includes:
将所述行业数据对齐到所述账户分类体系中进行词汇融合得到基础词语；Align the industry data into the account classification system and perform vocabulary fusion to obtain basic words;
根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。The basic words are added to the primary knowledge graph according to the association relationship between the basic vocabulary, the feature description sentence, and the plurality of preset accounts.
在其中一个实施例中，在所述根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，之后还包括：In one embodiment, after determining the relationship between the multiple preset accounts and the feature description sentence according to the feature information, adding the multiple preset accounts to the primary account knowledge graph, and then adding the multiple preset accounts to the primary account knowledge graph. Also includes:
根据所述行业数据通过预设的词语挖掘模型得到基础词语，所述词语挖掘模型为通过数据的语义特征和所述账户分类体系训练得到；According to the industry data, basic words are obtained through a preset word mining model, and the word mining model is obtained by training the semantic features of the data and the account classification system;
根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。The basic words are added to the primary knowledge graph according to the association relationship between the basic vocabulary, the feature description sentence, and the plurality of preset accounts.
第二方面，本公开实施例还提供了一种账户推荐的装置。所述装置包括：In a second aspect, an embodiment of the present disclosure further provides an account recommendation device. The device includes:
获取模块，用于获取第一账户的特征信息，所述特征信息包括所述第一账户在注册时填写的兴趣信息以及所述第一账户的评论信息；an acquisition module, configured to acquire feature information of the first account, where the feature information includes interest information filled in by the first account during registration and comment information of the first account;
确定模块，用于根据所述特征信息，确定第二账户，其中，所述第二账户的特征信息与所述第一账户的特征信息的相似度在预设范围内；a determining module, configured to determine a second account according to the feature information, wherein the similarity between the feature information of the second account and the feature information of the first account is within a preset range;
推荐模块，用于从预设的账户知识图谱中确定与所述第二账户的特征描述语句相匹配的目标账户，将所述目标账户推荐至所述第一账户，其中，所述账户知识图谱包括账户与特征描述语句之间的关联关系。A recommendation module, configured to determine a target account matching the feature description sentence of the second account from a preset account knowledge graph, and recommend the target account to the first account, wherein the account knowledge graph Including the association between accounts and feature description statements.
第三方面，本公开实施例还提供了一种计算机设备。所述计算机设备包括存储器和处理器，所述存储器存储有计算机程序，所述处理器执行所述计算机程序时实现本公开实施例中任一项所述的方法的步骤。In a third aspect, an embodiment of the present disclosure further provides a computer device. The computer device includes a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in any one of the embodiments of the present disclosure are implemented.
第四方面，本公开实施例还提供了一种计算机可读存储介质。所述计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现本公开实施例中任一项所述的方法的步骤。In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium. The computer-readable storage medium has a computer program stored thereon, and when the computer program is executed by a processor, implements the steps of the method described in any one of the embodiments of the present disclosure.
第五方面，本公开实施例还提供了一种计算机程序产品。所述计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现本公开实施例中任一项所述的方法的步骤。In a fifth aspect, an embodiment of the present disclosure further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the embodiments of the present disclosure.
本公开实施例，首先获取第一账户的特征信息，根据所述特征信息确定与第一账户相似度较高的第二账户，根据所述第二账户的特征描述语句确定预设的账户知识图谱中的目标账户，将所述目标账户推荐至第一账户，从而能够通过第二账户和账户知识图谱，提高了账户推荐的多样性和新意，在第一账户的数据较少的情况下，也能根据第二账户进行账户推荐，避免了因数据较少无法进行个性化推荐的问题提高了个性化推荐的准确性，优化了推荐内容。In this embodiment of the present disclosure, firstly obtain feature information of a first account, determine a second account with a high similarity to the first account according to the feature information, and determine a preset account knowledge graph according to the feature description sentence of the second account In the target account, the target account is recommended to the first account, so that the second account and the account knowledge graph can improve the diversity and novelty of account recommendation. In the case of less data in the first account, the The account recommendation can be made according to the second account, which avoids the problem that the personalized recommendation cannot be made due to the lack of data, improves the accuracy of the personalized recommendation, and optimizes the recommended content.
附图说明Description of drawings
图1为一个实施例中账户推荐的方法的流程示意图；1 is a schematic flowchart of a method for account recommendation in one embodiment;
图2为一个实施例中账户知识图谱的确定方法的流程示意图；2 is a schematic flowchart of a method for determining an account knowledge graph in one embodiment;
图3为一个实施例中账户知识图谱的构建方法的流程示意图；3 is a schematic flowchart of a method for constructing an account knowledge graph in one embodiment;
图4为一个实施例中账户知识图谱的结构示意图；4 is a schematic structural diagram of an account knowledge graph in one embodiment;
图5为一个实施例中账户推荐的方法的流程示意图；5 is a schematic flowchart of a method for account recommendation in one embodiment;
图6为一个实施例中账户推荐的装置的结构框图；6 is a structural block diagram of an apparatus for account recommendation in one embodiment;
图7为一个实施例中计算机设备的内部结构图。FIG. 7 is a diagram of the internal structure of a computer device in one embodiment.
具体实施方式Detailed ways
为了使本公开实施例的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本公开实施例进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本公开实施例，并不用于限定本公开实施例。In order to make the objectives, technical solutions and advantages of the embodiments of the present disclosure more clear, the embodiments of the present disclosure will be described in further detail below with reference to the accompanying drawings and the embodiments. It should be understood that the specific embodiments described herein are only used to explain the embodiments of the present disclosure, and are not used to limit the embodiments of the present disclosure.
在一个实施例中，如图1所示，提供了一种文件的发送方法，本实施例以该方法应用于服务器进行举例说明，可以理解的是，该方法也可以应用于终端，还可以应用于包括终端和服务器的系统，并通过终端和服务器的交互实现。本实施例中，该方法包括以下步骤：In one embodiment, as shown in FIG. 1 , a method for sending a file is provided. In this embodiment, the method is applied to a server for illustration. It can be understood that this method can also be applied to a terminal, and can also be applied to It is used in a system including a terminal and a server, and is realized through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:
步骤S110，获取第一账户的特征信息，所述特征信息包括所述第一账户在注册时填写的兴趣信息以及所述第一账户的评论信息；Step S110, acquiring characteristic information of the first account, where the characteristic information includes interest information filled in by the first account during registration and comment information of the first account;
本公开实施例中，首先获取第一账户的特征信息，特征信息包括第一账户在进行注册时填写的兴趣信息以及第一账户的评论信息，其中，第一账户注册时填写的兴趣信息可以包括但不限于注册时勾选的感兴趣的领域或注册时填写的个人信息，第一账户的评论信息可以包括第一账户评论的信息和其他账户对第一账户的评论信息。在一个示例中，所述第一账户可以为系统的新用户，此时，所述第一账户在系统内没有评论信息，所述评论信息可以为获取到的第一账户在系统外评论的信息以及系统外其他账户对第一系统的评论信息。In the embodiment of the present disclosure, first obtain the characteristic information of the first account, the characteristic information includes the interest information filled in by the first account during registration and the comment information of the first account, wherein the interest information filled in when the first account is registered may include However, it is not limited to the fields of interest checked during registration or the personal information filled during registration. The comment information of the first account may include the comment information of the first account and the comment information of other accounts on the first account. In one example, the first account may be a new user of the system. At this time, the first account has no comment information in the system, and the comment information may be the obtained information of the first account commenting outside the system. and comments on the first system by other accounts outside the system.
步骤S120，根据所述特征信息，确定第二账户，其中，所述第二账户的特征信息与所述第一账户的特征信息的相似度在预设范围内；Step S120, determining a second account according to the feature information, wherein the similarity between the feature information of the second account and the feature information of the first account is within a preset range;
本公开实施例中，在获取到第一账户的特征信息后，将所述第一账户的特征信息与系统内的账户的特征信息进行比较，当比较得到的特征信息的相似度在预设范围内时，确定对应的账户为第二账户。其中，根据特征信息确定相似度的确定方法可以包括但不限于通过计算实体之间的距离得到相似度，距离的计算方式可以包括欧式距离、杰卡德相似系数、余弦相似度等。在一个示例中，所述预设范围通常为根据实际场景确定的相似度高于预设值的范围，其中，预设值为预先设置的一个较高的值，当两个账户的特征信息的相似度高于预设值，即位于预设范围的情况下，可以认为此时两个账户的相似度较高。在另一个示例中，第二账户可以为一个也可以为多个。In the embodiment of the present disclosure, after the characteristic information of the first account is obtained, the characteristic information of the first account is compared with the characteristic information of the account in the system, and when the similarity of the characteristic information obtained by the comparison is within a preset range When the corresponding account is determined to be the second account. The method for determining the similarity according to the feature information may include, but is not limited to, calculating the distance between entities to obtain the similarity, and the calculation method for the distance may include Euclidean distance, Jaccard similarity coefficient, cosine similarity, and the like. In an example, the preset range is generally a range in which the similarity determined according to the actual scene is higher than a preset value, wherein the preset value is a preset higher value, and when the feature information of the two accounts is different When the similarity is higher than the preset value, that is, in the preset range, it can be considered that the similarity between the two accounts is high at this time. In another example, the second account may be one or multiple.
步骤S130，从预设的账户知识图谱中确定与所述第二账户的特征描述语句相匹配的目标账户，将所述目标账户推荐至所述第一账户，其中，所述账户知识图谱包括账户与特征描述语句之间的关联关系。Step S130, determining a target account matching the feature description sentence of the second account from a preset account knowledge graph, and recommending the target account to the first account, wherein the account knowledge graph includes an account The association relationship with the feature description statement.
本公开实施例中，在确定第二账户之后，根据所述第二账户对应的特征描述语句在预设的账户知识图谱中获取与所述特征描述语句相匹配的目标账户，并将所述目标账户推荐至第一账户。其中，所述特征描述语句通常为事先根据第二账户在系统内的评论数据、第二账户在系统内的历史行为数据等数据得到的对应的特征描述语句，特征描述语句为预先设置的可以描述账户需求的短语。在预设的账户知识图谱中包含了账户与特征描述语句的关联关系，所以在获取到第二账户的特征描述语句后，可以在账户知识图谱中确定与所述第二账户的特征描述语句存在关联关系的账户，得到目标账户，此时的目标账户与第二账户的需求较为吻合，由于第二账户与第一账户的相似度较高，所以可以认为此时的目标账户与所述第一账户的需求也较为吻合，将所述目标账户推荐给第一账户。In this embodiment of the present disclosure, after the second account is determined, a target account matching the feature description statement is acquired in a preset account knowledge graph according to the feature description statement corresponding to the second account, and the target account is assigned to the target account. The account is recommended to the first account. Wherein, the feature description statement is usually a corresponding feature description statement obtained in advance according to the comment data of the second account in the system, the historical behavior data of the second account in the system, etc., and the feature description statement is a preset description that can be described Phrase for account requirements. The preset account knowledge graph contains the association relationship between the account and the feature description sentence, so after obtaining the feature description sentence of the second account, it can be determined in the account knowledge graph that the feature description sentence related to the second account exists. The target account is obtained from an account in an associated relationship. The target account at this time is more consistent with the requirements of the second account. Since the second account is highly similar to the first account, it can be considered that the target account at this time is the same as the first account. The requirements of the accounts are also relatively consistent, and the target account is recommended to the first account.
本公开实施例，首先获取第一账户的特征信息，根据所述特征信息确定与第一账户相似度较高的第二账户，根据所述第二账户的特征描述语句确定预设的账户知识图谱中的目标账户，将所述目标账户推荐至第一账户，从而能够通过第二账户和账户知识图谱，提高了账户推荐的多样性和新意，在第一账户的数据较少的情况下，也能根据第二账户进行账户推荐，避免了因数据较少无法进行个性化推荐的问题提高了个性化推荐的准确性，优化了推荐内容。In this embodiment of the present disclosure, firstly obtain feature information of a first account, determine a second account with a high similarity to the first account according to the feature information, and determine a preset account knowledge graph according to the feature description sentence of the second account In the target account, the target account is recommended to the first account, so that the second account and the account knowledge graph can improve the diversity and novelty of account recommendation. In the case of less data in the first account, the The account recommendation can be made according to the second account, which avoids the problem that the personalized recommendation cannot be made due to the lack of data, improves the accuracy of the personalized recommendation, and optimizes the recommended content.
在一个实施例中，如图2所示，所述账户知识图谱的确定方法，包括：In one embodiment, as shown in FIG. 2 , the method for determining the account knowledge graph includes:
步骤S210，获取账户分类体系及特征描述语句，所述分类体系为根据账户的行业数据按照预设的分类规则处理得到；Step S210, acquiring an account classification system and a feature description statement, where the classification system is obtained by processing according to the industry data of the account and according to a preset classification rule;
步骤S220，根据所述分类体系及所述特征描述语句之间的关联关系，生成初级账户知识图谱；Step S220, generating a primary account knowledge graph according to the classification system and the association relationship between the feature description sentences;
步骤S230，获取多个预设账户及所述多个预设账户的特征信息；Step S230, acquiring multiple preset accounts and feature information of the multiple preset accounts;
步骤S240，根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，形成账户知识图谱。Step S240: Determine the relationship between the multiple preset accounts and the feature description sentence according to the feature information, and add the multiple preset accounts to the primary account knowledge graph to form an account knowledge graph.
本公开实施例中，需要确定账户知识图谱，首先获取账户所处的行业数据，根据所述行业数据确定账户分类体系以及特征描述语句，其中，所述行业数据可以包括但不限于行业内的用户评论数据、行业相关的数据等，所述特征描述语句为可以表达用户需求的短语，所述分类体系通常为多级分类体系。对行业数据按照预设的分类规则进行处理，得到账户分类体系。在一个示例中，分类规则可以由人工按照行业实际场景预先设置，也可以为对行业数据进行挖掘得到的分类规则。根据所述分类体系和所述特征描述语句之间的关联关系，生成一个网络体系，即初级账户知识图谱，其中，分类体系和特征描述语句之间的关联关系通常为根据特征描述语句的语义和分类体系之间的关系确定得到。然后，获取多个预设账户以及他们对应的特征信息，其中，多个预设账户为在系统中事先确定的能够提供服务的账户。根据所述预设账户的特征信息语义与所述特征描述语句语义之间的关系，建立所述多个预设账户与所述初级账户知识图谱之间的关联关系，形成账户知识图谱。In the embodiment of the present disclosure, it is necessary to determine the account knowledge graph, first obtain the industry data in which the account is located, and determine the account classification system and feature description sentences according to the industry data, wherein the industry data may include but not limited to users in the industry Comment data, industry-related data, etc., the feature description sentences are phrases that can express user needs, and the classification system is usually a multi-level classification system. The industry data is processed according to the preset classification rules to obtain the account classification system. In an example, the classification rules may be preset manually according to the actual industry scenarios, or may be classification rules obtained by mining industry data. According to the association relationship between the classification system and the feature description sentence, a network system is generated, that is, the primary account knowledge graph, wherein the association relationship between the classification system and the feature description sentence is usually based on the semantics and characteristics of the feature description sentence. The relationship between the classification systems is determined. Then, multiple preset accounts and their corresponding feature information are acquired, wherein the multiple preset accounts are pre-determined accounts that can provide services in the system. According to the relationship between the semantics of the feature information of the preset accounts and the semantics of the feature description sentence, the association relationship between the multiple preset accounts and the knowledge graph of the primary account is established to form an account knowledge graph.
本公开实施例，通过行业数据确定分类体系和能够体现用户需求的特征描述语句，并获取预设账户以及对应的特征信息，根据所述特征描述语句和所述分类体系之间的关联关系，所述预设账户与所述特征描述语句之间的关联关系，形成账户知识图谱，从而能够根据用户需求确定对应的目标账户，由于账户知识图谱中存在预设账户与特征描述语句的关联关系，提高了后续账户推荐的多样性。In this embodiment of the present disclosure, a classification system and feature description statements that can reflect user needs are determined through industry data, and preset accounts and corresponding feature information are acquired. According to the association between the feature description statements and the classification system, the The relationship between the preset account and the feature description sentence is formed to form an account knowledge graph, so that the corresponding target account can be determined according to user needs. The diversity of follow-up account recommendations.
在一个实施例中，所述特征描述语句的确定方式，包括：In one embodiment, the method for determining the feature description statement includes:
从所述行业数据中获取多个候选短语；obtain a plurality of candidate phrases from the industry data;
利用判别模型从所述多个候选短语确定特征描述语句，其中，所述判别模型为根据候选短语和候选短语对应的详细说明信息训练得到。The feature description sentences are determined from the plurality of candidate phrases by using a discriminant model, wherein the discriminant model is obtained by training according to the candidate phrases and the detailed description information corresponding to the candidate phrases.
本公开实施例中，获取到账户所在的行业数据之后，对所述行业数据进行挖掘得到多个候选短语。在一个示例中，候选短语的生成方式包括但不限于从文本语料(行业数据)中去挖掘可能的短语，通过爬虫在第三方的行业论坛、资讯的等网站爬取大规模的语料，在此基础上进行挖掘。语料包括行业内的日志、描述、评论等。在得到多个候选短语后，根据预设的判别模型对所述多个候选短语判断，根据判断的结果确定特征描述语句，其中，判别模型整体是一个Wide&Deep的结构，在Deep侧，利用字级别和词级别的BiLSTM来提取特征，同时对于词级别的输入，加入一些词性特征如POS tag(词性标签)和NER label(实体识别标签)等。在Wide侧，主要计算一些统计特征，包括了BERT语言模型产出的ppl值。最后，通过一个全连接层得到最终衡量一个候选短语是否符合特征描述语句要求的分数。其中，符合标准的特征描述语句通常需要有相关度，即和行业中的某个属性产生关联、指向明确、无错别字、语句通顺。In the embodiment of the present disclosure, after acquiring the industry data where the account is located, the industry data is mined to obtain a plurality of candidate phrases. In one example, the method of generating candidate phrases includes, but is not limited to, mining possible phrases from text corpus (industry data), and crawling large-scale corpora from third-party industry forums, news websites and other websites through crawler. based on excavation. The corpus includes logs, descriptions, comments, etc. in the industry. After a plurality of candidate phrases are obtained, the plurality of candidate phrases are judged according to a preset discriminant model, and feature description sentences are determined according to the judgment result, wherein the discriminant model as a whole is a Wide&Deep structure, and on the Deep side, the word level is used. and word-level BiLSTM to extract features, and for word-level input, add some part-of-speech features such as POS tag (part-of-speech tag) and NER label (entity recognition tag). On the Wide side, some statistical features are mainly calculated, including the ppl value produced by the BERT language model. Finally, a fully connected layer is used to obtain the final score that measures whether a candidate phrase meets the requirements of the feature description sentence. Among them, the characteristic description sentences that meet the standard usually need to have relevance, that is, they are related to a certain attribute in the industry, have clear directions, have no typos, and have smooth sentences.
本公开实施例，从获取到的行业数据中确定多个候选短语并从多个候选短语中确定特征描述语句，从而能够实现将用户需求转换成特征描述语句，提高了账户个性化推荐时的准确性，从而能够提高系统内的交易量。In this embodiment of the present disclosure, multiple candidate phrases are determined from the acquired industry data, and feature description sentences are determined from the multiple candidate phrases, so that user requirements can be converted into feature description sentences, and the accuracy of account personalized recommendation is improved. , which can increase the transaction volume within the system.
在一个实施例中，所述从所述行业数据中获取多个候选短语，还包括：In one embodiment, the obtaining a plurality of candidate phrases from the industry data further includes:
对所述账户的行业数据进行分词处理，得到词语；Perform word segmentation processing on the industry data of the account to obtain words;
根据所述词语的语义按照预设的组合规则对所述词语进行组合，得到多个候选短语。According to the semantics of the words, the words are combined according to a preset combination rule to obtain a plurality of candidate phrases.
本公开实施例中，从行业数据中获取候选短语时，可以通过对所述行业数据进行分词处理得到多个词语，根据所述多个词语的语义对其进行组合得到的多个短语为候选短语，其中，在对多个词语进行组合时，可以为按照预先设置的组合规则进行组合，也可以为根据训练得到的模型进行组合。In this embodiment of the present disclosure, when candidate phrases are obtained from industry data, multiple words may be obtained by performing word segmentation on the industry data, and multiple phrases obtained by combining the multiple words according to their semantics are candidate phrases , wherein, when combining multiple words, the combination may be performed according to a preset combination rule, or the combination may be performed according to a model obtained by training.
本公开实施例，通过对账户的行业数据进行分词处理得到词语，按照预设的组合规则对所述词语进行组合得到多个候选短语可以丰富候选短语，使得最终得到的特征描述语句更为全面，从而进一步提高了后续账号个性化推荐的准确性。In this embodiment of the present disclosure, words are obtained by segmenting the industry data of an account, and combining the words according to a preset combination rule to obtain multiple candidate phrases can enrich the candidate phrases, so that the finally obtained feature description sentences are more comprehensive, This further improves the accuracy of the follow-up account personalized recommendation.
在一个实施例中，在所述根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，之后还包括：In one embodiment, after determining the relationship between the multiple preset accounts and the feature description sentence according to the feature information, adding the multiple preset accounts to the primary account knowledge graph, and then adding the multiple preset accounts to the primary account knowledge graph. include:
将所述行业数据对齐到所述账户分类体系中进行词汇融合得到基础词语；Align the industry data into the account classification system and perform vocabulary fusion to obtain basic words;
根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。The basic words are added to the primary knowledge graph according to the association relationship between the basic vocabulary, the feature description sentence, and the plurality of preset accounts.
本公开实施例中，在构建账户知识图谱时，获取到行业数据后，将所述行业数据对齐到账户分类体系中进行词汇融合，得到多个词语，即为基础词语。在一个示例中，通常采用规则和人工映射相结合的方式将所述行业数据对齐到账户分类体系中进行词汇的融合，并通过预设的判别模型对所述得到的词语进行筛选，筛选后的词语即为基础词语。根据所述基础词语的语义和所述特征描述语句的语义确定基础词语和特征描述语句之间的关联关系，并根据所述基础词语与所述特征描述语句之间的关联关系、基础词语与所述分类体系之间的关联关系将所述基础词语加入到得到的初级知识图谱中。在一个示例中，将基础词语分配到分类体系中的方法通常可以抽象成为一个上下位关系发现的过程：给定一个下位词，在词表中找到其可能的上位词，其中，可以采用基于pattern的无监督方法和基于projection learning的监督方法两种方式结合来完成上下位关系的构建。In the embodiment of the present disclosure, when building an account knowledge graph, after obtaining industry data, the industry data is aligned into the account classification system for vocabulary fusion, and multiple words are obtained, which are basic words. In an example, a combination of rules and manual mapping is usually used to align the industry data into the account classification system for vocabulary fusion, and the obtained words are screened through a preset discriminant model. Words are basic words. The relationship between the basic word and the feature description statement is determined according to the semantics of the basic word and the feature description statement, and the relationship between the basic word and the feature description statement, the relationship between the basic word and the feature description statement According to the association between the classification systems, the basic words are added to the obtained primary knowledge graph. In one example, the method of assigning basic words to a taxonomy can often be abstracted into a process of discovery of hypernyms: given a hyponym, find its possible hypernyms in the vocabulary, which can be based on pattern The unsupervised method and the supervised method based on projection learning are combined to complete the construction of the upper and lower relationship.
本公开实施例，通过行业数据和账户分类体系得到基础词语，并将基础词语加入到初级账户知识图谱中，能够丰富账户知识图谱的内容，使得知识图谱中的预设账户和基础词语、特征描述语句之间的联系更为详细具体，在后续进行账户的个性化推荐时，推荐内容更为多样准确。In this embodiment of the present disclosure, basic words are obtained through industry data and an account classification system, and the basic words are added to the primary account knowledge map, which can enrich the content of the account knowledge map, so that the preset accounts, basic words, and feature descriptions in the knowledge map The connection between the sentences is more detailed and specific, and the recommended content is more diverse and accurate in the subsequent personalized recommendation of the account.
在一个实施例中，在所述根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，之后还包括：In one embodiment, after determining the relationship between the multiple preset accounts and the feature description sentence according to the feature information, adding the multiple preset accounts to the primary account knowledge graph, and then adding the multiple preset accounts to the primary account knowledge graph. include:
根据所述行业数据通过预设的词语挖掘模型得到基础词语，所述词语挖掘模型为通过数据的语义特征和所述账户分类体系训练得到；According to the industry data, basic words are obtained through a preset word mining model, and the word mining model is obtained by training the semantic features of the data and the account classification system;
根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。The basic words are added to the primary knowledge graph according to the association relationship between the basic vocabulary, the feature description sentence, and the plurality of preset accounts.
本公开实施例中，构建账户知识图谱时，获取到账户所在的行业数据之后，在所述行业数据的基础上通过序列标注任务自动挖掘补充分类体系的词语，其中，挖掘的方式可以为通过预设的词语挖掘模型得到，词语挖掘模型为根据数据的语义和账户分类体系中的分类标签之间的关系训练得到。在一个示例中，在得到词语后，还会通过判别模型对词语进行筛选，得到筛选后的词语为基础词语。In the embodiment of the present disclosure, when building an account knowledge graph, after acquiring the industry data where the account is located, the words of the supplementary classification system are automatically mined through sequence labeling tasks on the basis of the industry data. The set word mining model is obtained, and the word mining model is trained according to the relationship between the semantics of the data and the classification labels in the account classification system. In one example, after the words are obtained, the words are also screened by the discriminant model, and the filtered words are obtained as the basic words.
根据所述基础词语的语义和所述特征描述语句的语义确定基础词语和特征描述语句之间的关联关系，并根据所述基础词语与所述特征描述语句之间的关联关系、基础词语与所述分类体系之间的关联关系将所述基础词语加入到得到的初级知识图谱中。The relationship between the basic word and the feature description statement is determined according to the semantics of the basic word and the feature description statement, and the relationship between the basic word and the feature description statement, the relationship between the basic word and the feature description statement According to the association between the classification systems, the basic words are added to the obtained primary knowledge graph.
本公开实施例，通过行业数据和账户分类体系进行词汇挖掘得到基础词语，并将基础词语加入到初级账户知识图谱中，能够丰富账户知识图谱的内容，使得知识图谱中的预设账户和基础词语、特征描述语句之间的联系更为详细具体，在后续进行账户的个性化推荐时，推荐内容更为多样准确。In the embodiment of the present disclosure, basic words are obtained by vocabulary mining through industry data and account classification system, and the basic words are added to the primary account knowledge graph, so that the content of the account knowledge graph can be enriched, and the preset accounts and basic words in the knowledge graph can be enriched. , The relationship between the feature description sentences is more detailed and specific, and the recommended content is more diverse and accurate in the subsequent personalized recommendation of the account.
图3是根据一示例性实施例示出的一种账户知识图谱的构建方法的流程示意图，参考图3所示，在游戏平台中，需要按照用户需求将对应的游戏大神推荐给用户，就需要构建以游戏大神作为预设账户的账户知识图谱。首先将游戏平台内外的行业数据(主要包含对不同游戏评论等数据)结合专家知识(主要是产品的知识累积)，通过人工规则定义多级分类体系。定义好分类体系后，先挖掘出基础词语，即原子层下的词语，再构建基础词语和分类体系中分类标签之间的上下位关系，其中，通常通过两种方式快速扩充分类下的基础词语：一种是主要采用规则+人工映射的方式将平台内外的行业数据对齐到分类体系进行词汇融合。另外一种是通过在大规模的语料上进行自动挖掘来补充分类下的词汇，可以采用基于BiLSTM+CRF的模型来挖掘发现分类下的新词。在某个一级分类下的基础词语挖掘到一定量后，继续将所有基础词语分到不同层次的类别中去，即给定一个下位词，在词表中找到其可能的上位词。主要采用基于pattern的无监督方法和基于projection learning的监督方法两种方式结合来完成上下位关系的构建。随后确定需求层中的特征描述语句，通常先生成大量的候选短语，然后用判别模型来过滤那些不满足标准的候选短语。候选生成通常有两种方式：一种是从文本语料，即平台内外的行业数据中去挖掘可能的短语，另一种是用词粒度的基础词语进行组合生成短语粒度的特征描述语句。通过语义匹配模型和文本数据增强等方法，将站内的大神、基础词语、特征描述语句进行融合，最终如图4所示，形成一个大神相关的账户知识图谱。FIG. 3 is a schematic flowchart of a method for constructing an account knowledge graph according to an exemplary embodiment. Referring to FIG. 3 , in the game platform, it is necessary to recommend the corresponding game god to the user according to the user's needs, and then it is necessary to construct Account knowledge graph with Game God as the default account. First, the industry data inside and outside the game platform (mainly including data such as reviews of different games) is combined with expert knowledge (mainly the accumulation of product knowledge), and a multi-level classification system is defined through manual rules. After defining the classification system, first dig out the basic words, that is, the words under the atomic layer, and then construct the upper and lower relationship between the basic words and the classification labels in the classification system. Among them, there are usually two ways to quickly expand the basic words under the classification. : One is to align the industry data inside and outside the platform to the classification system for vocabulary fusion mainly by means of rules + manual mapping. The other is to supplement the vocabulary under the classification through automatic mining on a large-scale corpus. The model based on BiLSTM+CRF can be used to mine and discover new words under the classification. After mining a certain amount of basic words under a certain first-level classification, continue to classify all basic words into different levels of categories, that is, given a hyponym, find its possible hypernym in the vocabulary. The construction of the upper and lower relationship is mainly completed by combining the pattern-based unsupervised method and the projection learning-based supervised method. Then determine the feature description sentences in the requirement layer, usually generate a large number of candidate phrases first, and then use the discriminative model to filter those candidate phrases that do not meet the criteria. There are usually two ways to generate candidates: one is to mine possible phrases from text corpus, that is, industry data inside and outside the platform, and the other is to combine basic words at word granularity to generate phrase-granular feature description sentences. Through methods such as semantic matching model and text data enhancement, the gods, basic words, and feature description sentences in the site are integrated, and finally, as shown in Figure 4, a knowledge map of accounts related to gods is formed.
图5是根据一示例性实施例示出的一种账户推荐的方法的流程示意图，参考图5所示，当新用户注册之后，根据新用户注册时的广告特征，如注册时勾选的感兴趣的领域等，以及新用户的平台外的数据，通过用户相似度算法计算得到平台中与该新用户最接近的老用户，获取该老用户置信度最高的特征，通过账户知识图谱确定一批大神推荐给新用户。Fig. 5 is a schematic flowchart of a method for recommending an account according to an exemplary embodiment. Referring to Fig. 5, after a new user is registered, according to the advertisement characteristics of the new user when registering, such as the interest selected during registration domain, etc., as well as the data outside the new user's platform, calculate the old user in the platform that is closest to the new user through the user similarity algorithm, obtain the feature with the highest confidence of the old user, and determine a group of great gods through the account knowledge graph. Recommended for new users.
应该理解的是，虽然附图中的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，附图中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。It should be understood that although the steps in the flowchart in the accompanying drawings are shown in sequence according to the arrows, these steps are not necessarily executed in the sequence shown by the arrows. Unless explicitly stated herein, the execution of these steps is not strictly limited to the order, and these steps may be performed in other orders. Moreover, at least a part of the steps in the drawings may include multiple steps or multiple stages, these steps or stages are not necessarily executed at the same time, but may be executed at different times, and the execution sequence of these steps or stages is also It does not have to be performed sequentially, but may be performed alternately or alternately with other steps or at least a portion of the steps or stages within the other steps.
基于同样的发明构思，本公开实施例还提供了一种用于实现上述所涉及的账户推荐的方法的账户推荐的装置。该装置所提供的解决问题的实现方案与上述方法中所记载的实现方案相似，故下面所提供的一个或多个账户推荐的装置实施例中的具体限定可以参见上文中对于账户推荐的方法的限定，在此不再赘述。Based on the same inventive concept, an embodiment of the present disclosure also provides an account recommendation device for implementing the above-mentioned account recommendation method. The implementation solution for solving the problem provided by the device is similar to the implementation solution described in the above method. Therefore, for the specific limitations in the embodiments of one or more account recommendation devices provided below, please refer to the above description of the account recommendation method. limitations, which are not repeated here.
在一个实施例中，如图6所示，提供了一种账户推荐的装置，包括：In one embodiment, as shown in FIG. 6, a device for account recommendation is provided, including:
获取模块610，用于获取第一账户的特征信息，所述特征信息包括所述第一账户在注册时填写的兴趣信息以及所述第一账户的评论信息；an obtaining
确定模块620，用于根据所述特征信息，确定第二账户，其中，所述第二账户的特征信息与所述第一账户的特征信息的相似度在预设范围内；A
推荐模块630，用于从预设的账户知识图谱中确定与所述第二账户的特征描述语句相匹配的目标账户，将所述目标账户推荐至所述第一账户，其中，所述账户知识图谱包括账户与特征描述语句之间的关联关系。A
在一个实施例中，所述账户知识图谱的确定模块，包括：In one embodiment, the determining module of the account knowledge graph includes:
第一获取模块，用于获取账户分类体系及特征描述语句，所述分类体系为根据账户的行业数据按照预设的分类规则处理得到；The first obtaining module is used to obtain the account classification system and the feature description statement, and the classification system is obtained by processing according to the industry data of the account according to the preset classification rules;
生成模块，用于根据所述分类体系及所述特征描述语句之间的关联关系，生成初级账户知识图谱；A generating module, configured to generate a primary account knowledge graph according to the classification system and the relationship between the feature description sentences;
第二获取模块，用于获取多个预设账户及所述多个预设账户的特征信息；a second acquisition module, configured to acquire multiple preset accounts and feature information of the multiple preset accounts;
确定模块，用于根据所述特征信息确定所述多个预设账户与所述特征描述语句的关系，在所述初级账户知识图谱上加入所述多个预设账户，形成账户知识图谱。A determination module, configured to determine the relationship between the multiple preset accounts and the feature description sentence according to the feature information, and add the multiple preset accounts to the primary account knowledge graph to form an account knowledge graph.
在一个实施例中，所述特征描述语句的确定模块，包括：In one embodiment, the determining module of the feature description statement includes:
获取模块，用于从所述行业数据中获取多个候选短语；an acquisition module for acquiring multiple candidate phrases from the industry data;
确定模块，用于利用判别模型从所述多个候选短语确定特征描述语句，其中，所述判别模型为根据候选短语和候选短语对应的详细说明信息训练得到。A determination module, configured to determine feature description sentences from the plurality of candidate phrases by using a discriminant model, wherein the discriminant model is obtained by training according to the candidate phrases and the detailed description information corresponding to the candidate phrases.
在一个实施例中，所述获取模块，还包括：In one embodiment, the obtaining module further includes:
分词模块，用于对所述账户的行业数据进行分词处理，得到词语；A word segmentation module, which is used to perform word segmentation processing on the industry data of the account to obtain words;
组合模块，用于根据所述词语的语义按照预设的组合规则对所述词语进行组合，得到多个候选短语。The combination module is configured to combine the words according to the semantics of the words and according to a preset combination rule to obtain a plurality of candidate phrases.
在一个实施例中，所述确定模块，还包括：In one embodiment, the determining module further includes:
融合模块，用于将所述行业数据对齐到所述账户分类体系中进行词汇融合得到基础词语；A fusion module, used for aligning the industry data into the account classification system to perform vocabulary fusion to obtain basic words;
加入模块，用于根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。An adding module is configured to add the basic words to the primary knowledge graph according to the basic vocabulary, the feature description sentence, and the association relationship between the multiple preset accounts.
在其中一个实施例中，所述确定模块，还包括：In one embodiment, the determining module further includes:
挖掘模块，用于根据所述行业数据通过预设的词语挖掘模型得到基础词语，所述词语挖掘模型为通过数据的语义特征和所述账户分类体系训练得到；a mining module, configured to obtain basic words through a preset word mining model according to the industry data, and the word mining model is obtained by training the semantic features of the data and the account classification system;
加入模块，用于根据所述基础词汇和所述特征描述语句、所述多个预设账户之间的关联关系，在所述初级知识图谱中加入所述基础词语。An adding module is configured to add the basic words to the primary knowledge graph according to the basic vocabulary, the feature description sentence, and the association relationship between the multiple preset accounts.
上述账户推荐的装置中的各个模块可全部或部分通过软件、硬件及其组合来实现。上述各模块可以硬件形式内嵌于或独立于计算机设备中的处理器中，也可以以软件形式存储于计算机设备中的存储器中，以便于处理器调用执行以上各个模块对应的操作。Each module in the above-mentioned account recommendation device may be implemented in whole or in part by software, hardware and combinations thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory in the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
在一个实施例中，提供了一种计算机设备，该计算机设备可以是服务器，其内部结构图可以如图7所示。该计算机设备包括通过系统总线连接的处理器、存储器和网络接口。其中，该计算机设备的处理器用于提供计算和控制能力。该计算机设备的存储器包括非易失性存储介质、内存储器。该非易失性存储介质存储有操作系统、计算机程序和数据库。该内存储器为非易失性存储介质中的操作系统和计算机程序的运行提供环境。该计算机设备的数据库用于存储账户数据和账户知识图谱数据。该计算机设备的网络接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种账户推荐的方法。In one embodiment, a computer device is provided, and the computer device can be a server, and its internal structure diagram can be as shown in FIG. 7 . The computer device includes a processor, memory, and a network interface connected by a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium, an internal memory. The nonvolatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the execution of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store account data and account knowledge graph data. The network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program, when executed by the processor, implements a method of account recommendation.
本领域技术人员可以理解，图7中示出的结构，仅仅是与本公开实施例方案相关的部分结构的框图，并不构成对本公开实施例方案所应用于其上的计算机设备的限定，具体的计算机设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。Those skilled in the art can understand that the structure shown in FIG. 7 is only a block diagram of a partial structure related to the solution of the embodiment of the present disclosure, and does not constitute a limitation on the computer equipment to which the solution of the embodiment of the present disclosure is applied. The computer device may include more or fewer components than those shown in the figures, or combine certain components, or have a different arrangement of components.
在一个实施例中，还提供了一种计算机设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现上述各方法实施例中的步骤。In one embodiment, a computer device is also provided, including a memory and a processor, where a computer program is stored in the memory, and the processor implements the steps in the foregoing method embodiments when the processor executes the computer program.
在一个实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现上述各方法实施例中的步骤。In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, implements the steps in the foregoing method embodiments.
在一个实施例中，提供了一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现上述各方法实施例中的步骤。In one embodiment, a computer program product is provided, including a computer program, which implements the steps in each of the foregoing method embodiments when the computer program is executed by a processor.
需要说明的是，本公开实施例所涉及的用户信息(包括但不限于用户设备信息、用户个人信息等)和数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)，均为经用户授权或者经过各方充分授权的信息和数据。It should be noted that user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the embodiments of the present disclosure , are all information and data authorized by the user or fully authorized by all parties.
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本公开实施例所提供的各实施例中所使用的对存储器、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存、光存储器、高密度嵌入式非易失性存储器、阻变存储器(ReRAM)、磁变存储器(Magnetoresistive Random Access Memory，MRAM)、铁电存储器(Ferroelectric Random Access Memory，FRAM)、相变存储器(Phase Change Memory，PCM)、石墨烯存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器等。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic RandomAccess Memory，DRAM)等。本公开实施例所提供的各实施例中所涉及的数据库可包括关系型数据库和非关系型数据库中至少一种。非关系型数据库可包括基于区块链的分布式数据库等，不限于此。本公开实施例所提供的各实施例中所涉及的处理器可为通用处理器、中央处理器、图形处理器、数字信号处理器、可编程逻辑器、基于量子计算的数据处理逻辑器等，不限于此。Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage In the medium, when the computer program is executed, it may include the processes of the above-mentioned method embodiments. Wherein, any reference to a memory, a database, or other media used in the various embodiments provided by the embodiments of the present disclosure may include at least one of a non-volatile memory and a volatile memory. Non-volatile memory may include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetic memory (Magnetoresistive) Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene memory, etc. Volatile memory may include random access memory (Random Access Memory, RAM) or external cache memory, and the like. By way of illustration and not limitation, the RAM may be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM). The database involved in each of the embodiments provided by the embodiments of the present disclosure may include at least one of a relational database and a non-relational database. The non-relational database may include a blockchain-based distributed database, etc., but is not limited thereto. The processor involved in each embodiment provided by the embodiments of the present disclosure may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, and the like, Not limited to this.
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described. However, as long as there is no contradiction in the combination of these technical features It is considered to be the range described in this specification.
以上所述实施例仅表达了本公开实施例的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本公开实施例专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本公开实施例构思的前提下，还可以做出若干变形和改进，这些都属于本公开实施例的保护范围。因此，本公开实施例的保护范围应以所附权利要求为准。The above-mentioned embodiments only represent several implementations of the embodiments of the present disclosure, and the descriptions thereof are relatively specific and detailed, but should not be construed as limitations on the patent scope of the embodiments of the present disclosure. It should be noted that for those skilled in the art, without departing from the concept of the embodiments of the present disclosure, several modifications and improvements can be made, which all belong to the protection scope of the embodiments of the present disclosure. Therefore, the protection scope of the embodiments of the present disclosure should be governed by the appended claims.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210247532.XACN114625964A (en) | 2022-03-14 | 2022-03-14 | Account recommendation method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210247532.XACN114625964A (en) | 2022-03-14 | 2022-03-14 | Account recommendation method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114625964Atrue CN114625964A (en) | 2022-06-14 |
Family
ID=81902036
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210247532.XAPendingCN114625964A (en) | 2022-03-14 | 2022-03-14 | Account recommendation method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114625964A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205089A (en)* | 2014-06-30 | 2015-12-30 | 邻客音公司 | Account Recommendations |
| CN111400616A (en)* | 2020-03-31 | 2020-07-10 | 北京达佳互联信息技术有限公司 | Account recommendation method and device |
| CN112100504A (en)* | 2020-11-03 | 2020-12-18 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and storage medium |
| CN112269918A (en)* | 2020-10-09 | 2021-01-26 | 北京达佳互联信息技术有限公司 | Information recommendation method, device, equipment and storage medium |
| CN114117058A (en)* | 2020-08-31 | 2022-03-01 | 北京达佳互联信息技术有限公司 | Determination method, device, electronic device and storage medium of account information |
- 2022
- 2022-03-14CNCN202210247532.XApatent/CN114625964A/enactivePending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105205089A (en)* | 2014-06-30 | 2015-12-30 | 邻客音公司 | Account Recommendations |
| CN111400616A (en)* | 2020-03-31 | 2020-07-10 | 北京达佳互联信息技术有限公司 | Account recommendation method and device |
| CN114117058A (en)* | 2020-08-31 | 2022-03-01 | 北京达佳互联信息技术有限公司 | Determination method, device, electronic device and storage medium of account information |
| CN112269918A (en)* | 2020-10-09 | 2021-01-26 | 北京达佳互联信息技术有限公司 | Information recommendation method, device, equipment and storage medium |
| CN112100504A (en)* | 2020-11-03 | 2020-12-18 | 北京达佳互联信息技术有限公司 | Content recommendation method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108288229B (en) | A method for building user portraits | |
| US11227118B2 (en) | Methods, devices, and systems for constructing intelligent knowledge base | |
| CN109376222B (en) | Question-answer matching degree calculation method, question-answer automatic matching method and device | |
| CN111444320B (en) | Text retrieval method and device, computer equipment and storage medium | |
| CN110276023B (en) | POI transition event discovery method, apparatus, computing device and medium | |
| KR20200094627A (en) | Method, apparatus, device and medium for determining text relevance | |
| CN110750649A (en) | Knowledge graph construction and intelligent response method, device, equipment and storage medium | |
| US11734322B2 (en) | Enhanced intent matching using keyword-based word mover's distance | |
| US20190340503A1 (en) | Search system for providing free-text problem-solution searching | |
| CN111522915A (en) | Extraction method, device and equipment of Chinese event and storage medium | |
| US10496751B2 (en) | Avoiding sentiment model overfitting in a machine language model | |
| US20110320490A1 (en) | Named entity database or mining rule database update apparatus and method using named entity database and mining rule merged ontology schema | |
| CN117931858B (en) | Data query method, device, computer equipment and storage medium | |
| CN111125318B (en) | Method for improving knowledge graph relation prediction performance based on sememe-semantic item information | |
| CN115210705A (en) | Vector embedding model for relational tables with invalid or equivalent values | |
| US12406150B2 (en) | Machine learning systems and methods for many-hop fact extraction and claim verification | |
| WO2025071803A1 (en) | Dynamic prompt creation for large language models | |
| KR102153259B1 (en) | Data domain recommendation method and method for constructing integrated data repository management system using recommended domain | |
| CN119166744A (en) | Intent identification method, device, equipment and storage medium for financial service applications | |
| CN110795562B (en) | Graph optimization method, device, terminal and storage medium | |
| CN116911801A (en) | Activity scheme generation method, device, equipment and storage medium | |
| CN114625964A (en) | Account recommendation method and device, computer equipment and storage medium | |
| CN117688123A (en) | Method and device for generating document structure tree | |
| CN115391542A (en) | Classification model training method, text classification method, device and equipment | |
| CN115329083A (en) | Document classification method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |