Social security expanded surface feature construction and personnel screening method and device, and storage medium

Abstract

The invention discloses a social security expansion feature construction method, a social security expansion feature personnel screening method, equipment and a storage medium, and the method comprises the steps: constructing a knowledge graph according to the source data of each personnel participating in social insurance, and screening a graph substructure where the personnel conforming to an expansion rule are located from the knowledge graph; the expansion rule is used for mining potential personnel participating in or upgrading social insurance; generating an original expansion feature for each node in the atlas substructure; aggregating the original expanded surface features of the same atlas substructure into target expanded surface features of the personnel; and under the same surface expansion rule, splicing the target surface expansion features into social security surface expansion features. The embodiment has good flexibility and expandability, can quickly adapt to business requirements in a market environment, provides strong reasoning and mining capability, improves the accuracy of screening target crowds with to-be-expanded surfaces, and provides more powerful support for insured surface expansion work.

Description

Technical Field

[0001] The present invention relates to the technical field of computer processing, and in particular to a construction of social security expansion features and a personnel screening method, device and storage medium thereof. Background Art

[0002] In the business scenario of expanding social security coverage, the current way to identify the target population to be expanded is mainly for experts to define screening rules, and developers to write corresponding codes based on the screening rules and use SQL (Structured Query Language) statements to query the database.

[0003] The screening rules involve subjective judgments by experts, which can easily introduce biases. The screening rules are relatively fixed, less flexible, and less compatible with the market environment. In addition, the data distribution of social security participants is relatively inverse, belonging to multi-source heterogeneous data with weak internal correlation. SQL statements mainly support exact matching or fuzzy queries, and have low reasoning and mining capabilities, which can easily lead to errors and omissions when screening the target population to be expanded. Summary of the invention

[0004] In view of this, the present invention provides a construction of social security coverage expansion features and a personnel screening method, device and storage medium thereof, so as to improve the accuracy of screening the target population to be expanded.

[0005] A first aspect of the present invention provides a method for constructing social security coverage expansion features, comprising:

[0006] Construct a knowledge graph based on the source data of each person's participation in social insurance;

[0007] Filtering the graph substructure containing the personnel meeting the expansion rule from the knowledge graph; the expansion rule is used to mine the personnel who are potential participants or upgrade social insurance;

[0008] Generating original expanded surface features for each node in the graph substructure;

[0009] Aggregating the original face-expanding features of the same graph substructure into target face-expanding features of the person;

[0010] Under the same expansion rule, the target expansion features are spliced ​​into social security expansion features.

[0011] A second aspect of the present invention provides a method for screening personnel for social security expansion, comprising:

[0012] Build a knowledge graph based on the source data of each person's social insurance participation;

[0013] Filter the graph substructure where the newly added person is located from the knowledge graph;

[0014] Generating original expanded surface features for each node in the graph substructure;

[0015] Aggregating the original face-expanding features of the same graph substructure into target face-expanding features of the person;

[0016] Loading multiple social security expansion features constructed according to the method described in the first aspect;

[0017] When the target coverage expansion feature successfully matches any of the social security coverage expansion features, the newly added person is determined as a potential person to participate in or upgrade social insurance.

[0018] A third aspect of the present invention provides a device for constructing social security coverage expansion features, comprising:

[0019] The knowledge graph construction module is used to construct a knowledge graph based on the source data of each person participating in social insurance;

[0020] An expansion structure screening module is used to screen the graph substructures of the personnel that meet the expansion rules from the knowledge graph; the expansion rules are used to mine potential personnel who participate in or upgrade social insurance;

[0021] An original surface expansion feature generation module, used to generate original surface expansion features for each node in the graph substructure;

[0022] A target face expansion feature aggregation module, used for aggregating the original face expansion features of the same graph substructure into the target face expansion features of the person;

[0023] The social security expansion feature splicing module is used to splice the target expansion features into social security expansion features under the same expansion rules.

[0024] A fourth aspect of the present invention provides a device for screening people covered by social security, comprising:

[0025] The knowledge graph construction module is used to construct a knowledge graph based on the source data of each person participating in social insurance;

[0026] A new structure screening module is added to screen the graph substructure containing the newly added personnel from the knowledge graph;

[0027] An original surface expansion feature generation module, used to generate original surface expansion features for each node in the graph substructure;

[0028] A target face expansion feature aggregation module, used for aggregating the original face expansion features of the same graph substructure into the target face expansion features of the person;

[0029] A social security coverage expansion feature loading module, used to load multiple social security coverage expansion features constructed according to the method described in the first aspect;

[0030] The module for determining the personnel to be expanded is used to determine that the newly added personnel are potential personnel for participating in or upgrading social insurance when the target expansion feature successfully matches any of the social insurance expansion features.

[0031] A fifth aspect of the present invention provides an electronic device, the electronic device comprising:

[0032] at least one processor; and

[0033] a memory communicatively connected to the at least one processor; wherein,

[0034] The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for constructing social security coverage expansion features as described in the first aspect above or the method for screening social security coverage expansion personnel as described in the second aspect above.

[0035] The sixth aspect of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for constructing the social security coverage expansion features as described in the first aspect above or the method for screening social security coverage expansion personnel as described in the second aspect above.

[0036] The seventh aspect of the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the method for constructing the social security coverage expansion features as described in the first aspect above or the method for screening social security coverage expansion personnel as described in the second aspect above.

[0037] In this embodiment, a knowledge graph is constructed based on the source data of each person participating in social insurance, and the graph substructure where the persons who meet the expansion rules are located is selected from the knowledge graph; the expansion rules are used to mine potential persons who participate in or upgrade social insurance; the original expansion features are generated for each node in the graph substructure; the original expansion features of the same graph substructure are aggregated into the target expansion features of the person; under the same expansion rules, the target expansion features are spliced ​​into social insurance expansion features. This embodiment uses the knowledge graph to promote the sharing of multi-source heterogeneous data, aggregates multi-source heterogeneous data into a graph structure, so that the originally isolated source data can be interconnected to form a valuable information network, and performs association analysis on the knowledge graph, which can accurately identify the persons under each expansion rule, thereby constructing a generalized social insurance expansion feature, reducing the deviation of expert subjective judgment, and providing a data basis for screening the target population for expansion, so that it has good flexibility and scalability, can quickly adapt to business needs in the market environment, provide strong reasoning and mining capabilities, improve the accuracy of screening the target population to be expanded, and provide more powerful support for the work of insurance expansion.

[0038] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0040] Figure 1 It is a flow chart of a method for constructing social security coverage expansion features provided by an embodiment of the present invention.

[0041] Figure 2 It is a flow chart of a method for screening personnel for social security expansion provided by an embodiment of the present invention.

[0042] Figure 3 It is a structural schematic diagram of a device for constructing social security coverage expansion features provided by an embodiment of the present invention.

[0043] Figure 4 It is a structural schematic diagram of a device for screening people for social security expansion provided by an embodiment of the present invention.

[0044] Figure 5 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0046] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present invention described herein can cover sequential implementations other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0047] See also Figure 1 , shows a flow chart of a method for constructing social security coverage extension features provided by an embodiment of the present invention, the method can be executed by a device for constructing social security coverage extension features, the device for constructing social security coverage extension features can be implemented in the form of hardware and / or software, and the device for constructing social security coverage extension features can be configured in an electronic device. Figure 1 As shown, the method includes:

[0048] Step 101: Build a knowledge graph based on the source data of each person participating in social insurance.

[0049] In actual application, the full name of social security is social insurance. The business scenario for expanding the coverage of social security is to screen out people who should participate in social security but have not yet participated (i.e., potential participants in social insurance) and people who have participated in social insurance but can develop into higher-level insured persons (i.e., potential upgraded social insurance participants) based on the existing participants in social insurance and their insurance status, and take appropriate measures to incorporate these people into the social security system and expand the coverage of social security.

[0050] In this embodiment, the source data of each person's participation in social insurance can be read from a designated database.

[0051] Exemplarily, the source data includes at least one of the following:

[0052] 1. Basic data of personnel

[0053] The basic data of personnel refers to the basic situation of personnel, including name, gender, age, household registration address, ID number, etc.

[0054] 2. Insurance data of personnel

[0055] The insurance data of personnel indicates the situation of personnel participating in social insurance, including the name of the insured person, the ID number of the insured person, the type of insurance, the time of insurance, the payment situation, the information of personnel who have interrupted the insurance, etc.

[0056] 3. Employment data of personnel

[0057] The employment data of personnel indicates the employment status of personnel, including the name of the employed person, the ID number of the employed person, the employment unit, the type of occupation, the employment time, etc.

[0058] 4. Basic data of the enterprise

[0059] The basic data of an enterprise refers to the basic situation of the enterprise, including the company name, registered address, business scope, number of employees, etc.

[0060] Of course, the above-mentioned source data of social insurance participation is only used as an example. When implementing this embodiment, source data of other people participating in social insurance participation can be set according to actual conditions, and this embodiment does not limit this. In addition, in addition to the above-mentioned source data of social insurance participation, those skilled in the art can also use source data of other people participating in social insurance participation according to actual needs, and this embodiment does not limit this.

[0061] For the source data of each person participating in social insurance, appropriate methods can be used to clean the data to improve the quality of the source data of each person participating in social insurance.

[0062] Generally speaking, the method of data cleaning for the source data of each person participating in social insurance includes processing duplicate values, missing values, and abnormal values.

[0063] For duplicate values ​​(i.e., there are multiple duplicate source data), the source data with the highest confidence (i.e., the most authoritative) can be selected based on factors such as the source.

[0064] For missing values ​​(i.e., there are missing fields in the source data), if the field is judged to have little impact, it can be deleted directly. If it has a great impact on subsequent use, the data of the same field in the source data from other channels will be filled into the field of the current source data.

[0065] For abnormal values ​​(i.e., abnormal values ​​of source data), such as values ​​beyond the normal range or illogical values, the source data with the highest confidence level (i.e., the most authoritative) can be selected for correction based on factors such as the source. If the value of the source data is judged to be noise data, the value of the source data is deleted.

[0066] Of course, the above data cleaning methods are only examples. When implementing this embodiment, other data cleaning methods can be set according to actual conditions, and this embodiment does not limit this. In addition, in addition to the above data cleaning methods, those skilled in the art can also use other data cleaning methods according to actual needs, and this embodiment does not limit this.

[0067] The source data of each person participating in social insurance can be structured and a knowledge graph can be constructed. The knowledge graph is a semantic network that reveals the relationship between entities and can be used to represent the insurance relationship.

[0068] In one embodiment of the present invention, step 101 may include the following steps:

[0069] Step 1011: Filter entities related to participation in social insurance from source data of each person's participation in social insurance.

[0070] In this embodiment, the source data of each person's participation in social insurance can be traversed to define the main unit (ie, entity) type, thereby filtering out entities related to participation in social insurance.

[0071] Exemplarily, entities include persons, enterprises, insurance records, occupations, regions, and family relationships.

[0072] In this example, the knowledge graph is constructed by developing a relationship graph around the three major entities of people, enterprises, and government agencies. It mainly displays the basic information network of people, enterprises, and agencies. On this basis, the business data of the four major business segments of employment and entrepreneurship, social security, labor relations, and human resources and talents are combined to construct a knowledge graph serving the four major business areas.

[0073] Step 1012: Query the source data of each person's participation in social insurance for the relationship between the entity representing the person and other entities in terms of participation in social insurance.

[0074] In this embodiment, the source data of each person's participation in social insurance can be traversed to define the semantic relationship types between different entities. The entity representing the person is taken as the center, and the relationship between the entity representing the person and other entities except the entity representing the person in participating in social insurance can be queried.

[0075] Exemplarily, for entities such as persons, enterprises, insurance records, occupations, regions, and family relationships, the relationship between the entity representing persons and the entity representing enterprises is an employment relationship, the relationship between the entity representing persons and the entity representing insurance records is an insurance relationship, the relationship between the entity representing persons and the entity representing occupations is an engaged relationship, the relationship between the entity representing persons and the entity representing regions is a residence relationship and / or employment location relationship, and the relationship between the entity representing persons and the entity representing family relationships is a kinship relationship.

[0076] Step 1013: Query the parameters of each entity in participating in social insurance in the source data of each person participating in social insurance.

[0077] In this embodiment, the source data of each person's participation in social insurance can be traversed, from which the types of parameters can be defined, and the parameters of each entity in participating in social insurance can be queried.

[0078] Exemplarily, for entities such as personnel, enterprises, insurance records, occupations, regions, and family relationships, the parameters of personnel include at least one of name, gender, age, ID number, mobile phone number, and occupation; the parameters of enterprises include at least one of enterprise name, enterprise type, address, contact number, and number of insured persons; the parameters of insurance records include at least one of insured type, insurance time, and insurance payment details; the parameters of occupations include at least one of occupation name and occupation salary; the parameters of regions include address information of each level (such as province, city, district / county, etc.); and the parameters of family relationships include relationship types (such as parents, spouse, children, etc.).

[0079] Step 1014: define entities as nodes, parameters as attributes of nodes, and relationships as edges to construct a knowledge graph that characterizes the characteristics of participating in social insurance.

[0080] In the business scenario of expanding the coverage of social security, entities, relationships, and parameters are extracted from the source data of each person participating in social insurance as usable knowledge units, and a knowledge graph is constructed based on this to form a series of high-quality factual expressions, laying the foundation for the construction of the upper model layer.

[0081] The definition of the graph structure in the knowledge graph includes the definition of nodes and edges, where entities correspond to nodes of the graph structure. By way of example, for entities such as personnel, enterprises, insurance records, occupations, regions, and family relationships, the node types of the graph structure include nodes representing personnel, nodes representing enterprises, nodes representing insurance records, nodes representing occupations, nodes representing regions, nodes representing family relationships, etc.

[0082] When the node definition is completed, the properties of the node can be defined according to the personnel parameters.

[0083] In addition, relationships correspond to edges in the graph structure. Edges represent relationships between nodes, which are also relationships between entities. The main ones include employment relationships, insurance relationships, engagement relationships, residence relationships, employment location relationships, and kinship relationships.

[0084] In practical applications, columns can be used to represent entities in a table, and columns or combinations of columns can represent relationships. Corresponding conversion logic can be designed for each table to map the data in the rows and columns of the table to the nodes and edges in the graph structure.

[0085] Use programming languages ​​(such as Python, Java, etc.) to write the conversion logic into conversion code, and generate statements that can be recognized by the graph database. With the help of some third-party tool libraries, the conversion code is loaded to realize the conversion from table to graph structure, so as to build a knowledge graph in the business scenario of expanding social security coverage.

[0086] Of course, the nodes, relationships and attributes in the above-mentioned knowledge graph are only used as examples. When implementing this embodiment, nodes, relationships and attributes in other knowledge graphs can be set according to actual conditions, and this embodiment does not limit this. In addition, in addition to the nodes, relationships and attributes in the above-mentioned knowledge graph, those skilled in the art can also use nodes, relationships and attributes in other knowledge graphs according to actual needs, and this embodiment does not limit this.

[0087] Step 102: Filter the graph substructure containing the personnel meeting the expansion rules from the knowledge graph.

[0088] In this embodiment, one or more expansion rules can be dynamically formulated in advance based on the business scenario of expanding the coverage of social insurance, wherein the expansion rules are used to discover potential personnel who may participate in or upgrade social insurance.

[0089] According to the expansion rules, it is converted into a query strategy that is adapted to the knowledge graph (i.e., graph retrieval logic), and the query logic is used to filter out the corresponding structure from the knowledge graph to obtain the graph substructure.

[0090] In the business scenario of expanding the coverage of social security, the target of expansion is personnel, and the graph substructure is a local graph structure related to the nodes representing personnel.

[0091] Furthermore, persons who meet the expansion rules can be pushed to relevant experts for review to improve the accuracy of screening persons who meet the expansion rules, thereby improving the accuracy of social security expansion features.

[0092] In a specific implementation, the node where the person who meets the preset expansion rules is located can be queried in the knowledge graph as the central node.

[0093] Exemplarily, the expansion rule includes at least one of the following:

[0094] 1. Those who have participated in occupational injury insurance (especially occupational injury insurance for new business forms) but have not participated in basic pension insurance for enterprise employees.

[0095] 2. The applicant has participated in work-related injury insurance according to the project, but has not participated in the basic pension insurance for enterprise employees.

[0096] 3. Those who have registered for employment but have not participated in the enterprise employee pension insurance.

[0097] 4. Terminate participation in enterprise employee pension insurance.

[0098] 5. The insurance and payment level of urban and rural residents' basic medical insurance is greater than or equal to the preset threshold (such as 3600), and they have not participated in the enterprise employees' pension insurance.

[0099] 6. Individual business owners who have not participated in the basic pension insurance for enterprise employees.

[0100] Generally speaking, there are certain differences in the expansion objectives of various expansion rules. For example, the expansion rule of item 1 is to screen out those who participate in occupational injury insurance and whose insurance types do not include the basic pension insurance for enterprise employees. Among them, new business occupational injury insurance is an occupational injury insurance designed for practitioners of various new business forms. It aims to provide relevant practitioners with accidental risk protection due to sudden, non-disease accidents or other eligible circumstances during the work process, and they have the potential to participate in the basic pension insurance for enterprise employees.

[0101] Of course, the above expansion rules are only used as examples. When implementing this embodiment, other expansion rules can be set according to actual conditions, and this embodiment does not limit this. In addition, in addition to the above expansion rules, those skilled in the art can also adopt other expansion rules according to actual needs, and this embodiment does not limit this.

[0102] In the knowledge graph, other nodes adjacent to the central node (i.e., the first-order subgraph nodes centered on the central node) are searched as neighbor nodes. Neighbor nodes include nodes representing enterprises, nodes representing insurance records, nodes representing occupations, nodes representing regions, nodes representing family relationships, etc.

[0103] The central node, neighbor nodes, the attributes of the central node, the attributes of the neighbor nodes, and the relationship between the central node and the neighbor nodes are extracted to obtain the graph substructure.

[0104] Step 103: Generate original expanded surface features for each node in the graph substructure.

[0105] In this embodiment, for each graph substructure, feature engineering can be performed on each node in the graph substructure to convert it into a feature representation, which is recorded as the original expanded feature.

[0106] In the specific implementation, triple data (head, relation, tail) can be extracted from the historical knowledge graph, where head is the head, relation is the relationship, and tail is the tail. Both the head and the tail are entities with attributes. The triple data is used as a sample to train a translation model. The translation models may include TransE, TransH, TransR, TransD, etc.

[0107] Taking the TransR model as an example, the TransR model models entities and relations in two different spaces, namely, the entity space and multiple relation spaces (relationship-specific entity spaces), and performs transformations in the corresponding relation spaces.

[0108] When the construction of the graph substructure is completed, a preset translation model suitable for processing the knowledge graph can be loaded, and the entities corresponding to each node, the relationships corresponding to the edges connecting the nodes, and the attributes corresponding to the nodes can be extracted from the graph substructure as triple data. The triple data is input into the translation model for processing, thereby generating the original expanded features of each node.

[0109] Step 104: Aggregate the original expanded face features of the same graph substructure into the target expanded face features of the person.

[0110] In this embodiment, for each graph substructure, the original expanded features of each node in the same graph substructure can be aggregated into the features of the personnel mined under the corresponding expanded rules, which are recorded as target expanded features.

[0111] In the specific implementation, you can load pre-set graph neural networks suitable for processing knowledge graphs, such as Graph Convolution Networks (GCN), Graph Attention Networks (GAN), Graph Autoencoders (GA), Graph Generative Networks (GGN) and Graph Spatial-temporal Networks (GSN).

[0112] Among them, GCN includes improved models such as R-GCN (Modeling Relational Data with Graph Convolutional Networks, relational data modeling based on graph convolutional networks).

[0113] On the one hand, the graph substructure is converted into an adjacency matrix.

[0114] On the other hand, in the same graph substructure, a two-dimensional feature matrix is ​​constructed with the original expanded features of each node as rows.

[0115] On the other hand, in the same graph substructure, a relationship feature matrix is ​​constructed with the relationships between nodes as rows and the dimensions of the original expanded features of each node as columns.

[0116] When the relationship is an employment relationship, an insurance relationship, an engagement relationship, a residence relationship, an employment location relationship, or a kinship relationship, and the dimension of the original expanded feature is n, the relationship feature matrix is ​​a matrix with a dimension of 6×n.

[0117] The adjacency matrix, two-dimensional feature matrix and relationship feature matrix are input into the graph neural network for processing. For each node representing a person, the graph neural network will aggregate the information of all its neighboring nodes, and fuse this information with the original expanded surface features of the node representing the person itself, thereby generating the target expanded surface features of the person mined under the same expanded surface rule.

[0118] In this embodiment, the method of combining knowledge graph with graph neural network can automatically extract key information from massive data, build complex social relationship networks, and quickly generate target expansion features of personnel. This process greatly reduces manual intervention and improves the efficiency and accuracy of screening.

[0119] Step 105: Under the same expansion rule, the target expansion features are spliced ​​into social security expansion features.

[0120] In each graph substructure, there is a central node corresponding to the person who meets the expansion rules. Each central node will generate a corresponding target expansion feature after being processed by the graph neural network. These target expansion features can represent the graph relationship structure characteristics under the knowledge graph of the target population that meets the expansion rules to the greatest extent.

[0121] Traverse the target expansion features of all personnel, filter out the target expansion features of personnel mined under the same expansion rule, and then splice some or all of the target expansion features under the same expansion rule into new features, which are recorded as the social security expansion features of the same expansion rule.

[0122] Assuming that there are n people who meet a certain expansion rule, there are n graph substructures, corresponding to the central nodes of the n people and the first-order neighbor nodes of the n central nodes. The target expansion features of the n central nodes are represented as A1, A2, ..., A n , these 10 target expansion features are concatenated on the vector to form the social security expansion feature B that meets the same expansion rule.

[0123] In this embodiment, a knowledge graph is constructed based on the source data of each person participating in social insurance, and the graph substructure where the persons who meet the expansion rules are located is selected from the knowledge graph; the expansion rules are used to mine potential persons who participate in or upgrade social insurance; the original expansion features are generated for each node in the graph substructure; the original expansion features of the same graph substructure are aggregated into the target expansion features of the person; under the same expansion rules, the target expansion features are spliced ​​into social insurance expansion features. This embodiment uses the knowledge graph to promote the sharing of multi-source heterogeneous data, aggregates multi-source heterogeneous data into a graph structure, so that the originally isolated source data can be interconnected to form a valuable information network, and performs association analysis on the knowledge graph, which can accurately identify the persons under each expansion rule, thereby constructing a generalized social insurance expansion feature, reducing the deviation of expert subjective judgment, and providing a data basis for screening the target population for expansion, so that it has good flexibility and scalability, can quickly adapt to business needs in the market environment, provide strong reasoning and mining capabilities, improve the accuracy of screening the target population to be expanded, and provide more powerful support for the work of insurance expansion.

[0124] See also Figure 2 , shows a flow chart of a method for screening social security personnel expansion provided by an embodiment of the present invention, the method can be performed by a screening device for social security personnel expansion, the device for constructing social security coverage features can be implemented in the form of hardware and / or software, and the screening device for social security personnel expansion can be configured in an electronic device. Figure 2 As shown, the method includes:

[0125] Step 201: construct a knowledge graph based on the source data of each person participating in social insurance.

[0126] In this embodiment, the source data of each person's participation in social insurance can be read from a designated database regularly or irregularly to construct a knowledge graph.

[0127] When the knowledge graph was constructed last time, the source data of relevant personnel participating in social insurance had been prepared. When the knowledge graph is constructed this time, the source data of newly added personnel participating in social insurance can be added on the basis of the original source data of relevant personnel participating in social insurance, so as to quickly prepare the source data of each person participating in social insurance.

[0128] In the specific implementation, entities related to participation in social insurance are screened in the source data of each person participating in social insurance; the relationship between the entity representing the person and other entities in participating in social insurance is queried in the source data of each person participating in social insurance; the parameters of each entity in participating in social insurance are queried in the source data of each person participating in social insurance; entities are defined as nodes, parameters are defined as attributes of nodes, and relationships are defined as edges, so as to construct a knowledge graph that characterizes the characteristics of participating in social insurance.

[0129] Exemplarily, the source data includes at least one of the following:

[0130] Basic data of personnel, insurance data of personnel, employment data of personnel, and basic data of enterprises.

[0131] Exemplarily, the relationship between an entity representing a person and an entity representing an enterprise is an employment relationship, the relationship between an entity representing a person and an entity representing an insurance record is an insurance relationship, the relationship between an entity representing a person and an entity representing an occupation is an engagement relationship, the relationship between an entity representing a person and an entity representing a region is a residence relationship and / or an employment location relationship, and the relationship between an entity representing a person and an entity representing a family relationship is a kinship relationship;

[0132] Exemplarily, the parameters of personnel include at least one of name, gender, age, ID number, mobile phone number, and occupation; the parameters of enterprises include at least one of enterprise name, enterprise type, address, contact number, and number of insured persons; the parameters of insurance records include at least one of insurance type, insurance time, and insurance payment details; the parameters of occupation include at least one of occupation name and occupation salary; the parameters of regions include address information of each level; and the parameters of family relationships include relationship type.

[0133] Step 202: Filter the graph substructure containing the newly added personnel from the knowledge graph.

[0134] In this embodiment, a query strategy (i.e., graph retrieval logic) adapted to the knowledge graph can be formulated based on the information of the newly added personnel (such as ID, etc.), and the query logic can be used to filter out the corresponding structure from the knowledge graph to obtain the graph substructure where the newly added personnel are located.

[0135] In the specific implementation, in the knowledge graph, the node where the newly added person is located is queried as the central node; in the knowledge graph, other nodes adjacent to the central node are searched as neighbor nodes; the central node, neighbor nodes, the attributes of the central node, the attributes of the neighbor nodes, and the relationship between the central node and the neighbor nodes are extracted to obtain the graph substructure.

[0136] Step 203: Generate original expanded surface features for each node in the graph substructure.

[0137] In this embodiment, for the graph substructure of the newly added personnel, feature engineering can be performed on each node in the high graph substructure to convert it into a feature representation, which is recorded as the original expanded feature.

[0138] In the specific implementation, the preset translation model is loaded; the entities corresponding to each node, the relationships corresponding to the edges connecting the nodes, and the attributes corresponding to the nodes are extracted from the graph substructure as triple data; the triple data is input into the translation model to generate the original expanded features of each node.

[0139] Step 204: Aggregate the original face-expanding features of the same graph substructure into the target face-expanding features of the person.

[0140] In this embodiment, for the graph substructure of the newly added personnel, the original expanded features of each node in the same graph substructure can be aggregated into features of the newly added personnel, which are recorded as target expanded features.

[0141] In the specific implementation, the preset graph neural network is loaded; the graph substructure is converted into an adjacency matrix; in the same graph substructure, a two-dimensional feature matrix is ​​constructed with the original expanded features of each node as rows; in the same graph substructure, a relationship feature matrix is ​​constructed with the relationship between each node as rows and the dimensions of the original expanded features of each node as columns; the adjacency matrix, the two-dimensional feature matrix and the relationship feature matrix are input into the graph neural network to generate the target expanded features of the personnel mined under the same expansion rule.

[0142] Step 205: Load multiple social security expansion features.

[0143] In this embodiment, social security coverage expansion features under multiple expansion rules constructed according to the construction method of social security coverage expansion features can be loaded.

[0144] Step 206: When the target coverage expansion feature successfully matches any social security coverage expansion feature, the newly added person is determined as a potential person to participate in or upgrade social insurance.

[0145] In this embodiment, the target expansion features of the newly added personnel may be matched with the social security expansion features under each expansion rule.

[0146] In the specific implementation, the similarity between the target expansion features of new personnel and the social security expansion features under various expansion rules can be calculated, such as Euclidean distance, cosine of angle, etc.

[0147] Taking the cosine of the angle as an example, the cosine of the angle between the target expansion characteristics of new personnel and the social security expansion characteristics under each expansion rule is expressed as follows:

[0148]

[0149] Among them, Cosine Similarity is the cosine of the angle, A is the target expansion feature of the new personnel, B is the social security expansion feature under a certain expansion rule, A·B is the dot product (inner product) between the target expansion feature of the new personnel and the social security expansion feature under a certain expansion rule, ||A|| is the modulus (length) of the target expansion feature of the new personnel, and ||B|| is the modulus (length) of the social security expansion feature under a certain expansion rule.

[0150] Assume that the target expansion feature A of the new personnel is an n-dimensional vector A = (A1, A2, ..., A n ), then its modulus is defined as

[0151] Assume that the social security coverage expansion feature B under a certain expansion rule is an n-dimensional vector B = (B1, B2, ..., B n ), then its modulus is defined as

[0152] If the similarity between the target expansion characteristics of the new personnel and the social security expansion characteristics under a certain expansion rule is greater than or equal to a preset threshold (such as 0.8), it means that the new personnel and the personnel under a certain expansion rule have a relatively similar relationship structure, and it can be considered that the target expansion characteristics of the new personnel and the social security expansion characteristics under the expansion rule are successfully matched.

[0153] If the similarity between the target expansion features of the new personnel and the social security expansion features under all expansion rules is less than the preset threshold (such as 0.8), it means that the new personnel and the personnel under all expansion rules do not have a relatively similar relationship structure. It can be considered that the target expansion features of the new personnel and the social security expansion features under all expansion rules fail to match.

[0154] When the target expansion feature successfully matches any social security expansion feature, it is inferred that the person has development potential, and further inferred that he or she belongs to the target population for expansion. At this time, the newly added person can be determined as a potential person to participate in or upgrade social insurance, assisting relevant business personnel in making expansion decisions.

[0155] This embodiment constructs a knowledge graph based on the source data of each person participating in social insurance; selects the graph substructure where the newly added person is located from the knowledge graph; generates original expansion features for each node in the graph substructure; aggregates the original expansion features of the same graph substructure into the target expansion features of the person; loads multiple social insurance expansion features; and when the target expansion feature successfully matches any social insurance expansion feature, determines that the newly added person is a potential person to participate in or upgrade social insurance. This embodiment uses the knowledge graph to promote the sharing of multi-source heterogeneous data, aggregates multi-source heterogeneous data into a graph structure, so that the originally isolated source data can be interconnected to form a valuable information network, and performs association analysis on the knowledge graph, which can accurately identify the target population that meets the expansion rules, reduce the deviation of expert subjective judgment, and make it have good flexibility and scalability, can quickly adapt to business needs in the market environment, provide strong reasoning and mining capabilities, improve the accuracy of screening the target population to be expanded, and provide more powerful support for the work of expanding insurance coverage.

[0156] See also Figure 3 , which shows a schematic diagram of the structure of a device for constructing social security coverage extension features provided by an embodiment of the present invention. Figure 3 As shown, the device comprises:

[0157] The knowledge graph construction module 301 is used to construct a knowledge graph based on the source data of each person participating in social insurance;

[0158] The expansion structure screening module 302 is used to screen the graph substructures of the personnel that meet the expansion rules from the knowledge graph; the expansion rules are used to mine potential personnel who participate in or upgrade social insurance;

[0159] The original surface expansion feature generation module 303 is used to generate the original surface expansion feature for each node in the graph substructure;

[0160] A target face-expanding feature aggregation module 304 is used to aggregate the original face-expanding features of the same graph substructure into the target face-expanding features of the person;

[0161] The social security expansion feature splicing module 305 is used to splice the target expansion features into social security expansion features under the same expansion rule.

[0162] In one embodiment of the present invention, the knowledge graph construction module 301 includes:

[0163] An entity screening module is used to screen entities related to social insurance participation in the source data of each person participating in social insurance;

[0164] A relationship query module, used to query the relationship between the entity representing the person and other entities in participating in social insurance in the source data of each person participating in social insurance;

[0165] A parameter query module, used to query the parameters of each entity in participating in social insurance in the source data of each person participating in social insurance;

[0166] The knowledge graph generation module is used to define the entities as nodes, the parameters as attributes of the nodes, and the relationships as edges, so as to construct a knowledge graph representing the characteristics of participating in social insurance.

[0167] In one embodiment of the present invention, the source data includes at least one of the following:

[0168] The basic data of the personnel, the insurance data of the personnel, the employment data of the personnel, and the basic data of the enterprise;

[0169] The relationship between the entity representing a person and the entity representing an enterprise is an employment relationship, the relationship between the entity representing a person and the entity representing insurance records is an insurance relationship, the relationship between the entity representing a person and the entity representing an occupation is an engagement relationship, the relationship between the entity representing a person and the entity representing a region is a residence relationship and / or an employment location relationship, and the relationship between the entity representing a person and the entity representing a family relationship is a kinship relationship;

[0170] The parameters of the personnel include at least one of name, gender, age, ID number, mobile phone number, and occupation; the parameters of the enterprise include at least one of enterprise name, enterprise type, address, contact number, and number of insured persons; the parameters of the insurance record include at least one of the insurance type, insurance time, and insurance payment details; the parameters of the occupation include at least one of the occupation name and occupation salary; the parameters of the region include address information of each level; and the parameters of the family relationship include relationship type.

[0171] In one embodiment of the present invention, the surface expansion structure screening module 302 includes:

[0172] A central node query module, used to query the node where the person who meets the preset expansion rule is located in the knowledge graph as the central node;

[0173] A neighbor node search module, used to search for other nodes adjacent to the central node in the knowledge graph as neighbor nodes;

[0174] The graph sub-node extraction sub-module is used to extract the central node, the neighboring nodes, the attributes of the central node, the attributes of the neighboring nodes, and the relationship between the central node and the neighboring nodes to obtain a graph sub-structure.

[0175] In one embodiment of the present invention, the expansion rule includes at least one of the following:

[0176] Those who have participated in occupational injury insurance but have not participated in basic pension insurance for enterprise employees;

[0177] Participated in work-related injury insurance according to the project, but not participated in the basic pension insurance for enterprise employees;

[0178] Those who have registered for employment but have not participated in the enterprise employee pension insurance;

[0179] Suspension of participation in enterprise employee pension insurance;

[0180] The participation and payment level of the basic medical insurance for urban and rural residents is greater than or equal to the preset threshold, and they have not participated in the enterprise employee pension insurance;

[0181] Belong to individual business owners and have not participated in the basic pension insurance for enterprise employees.

[0182] In one embodiment of the present invention, the original surface expansion feature generation module 303 includes:

[0183] The translation model loading module is used to load the preset translation model;

[0184] A triple data extraction module, used to extract the entity corresponding to each node, the relationship corresponding to the edge connected to the node and the attribute corresponding to the node in the graph substructure as triple data;

[0185] The feature translation module is used to input the triplet data into the translation model to generate the original expanded surface features of each node.

[0186] In one embodiment of the present invention, the target surface expansion feature aggregation module 304 includes:

[0187] The graph neural network loading module is used to load the preset graph neural network;

[0188] An adjacency matrix conversion module, used for converting the graph substructure into an adjacency matrix;

[0189] A two-dimensional feature matrix construction module is used to construct a two-dimensional feature matrix in the same graph substructure, taking the original expanded surface features of each node as rows;

[0190] A relationship feature matrix construction module is used to construct a relationship feature matrix in the same graph substructure, with the relationship between each node as the row and the dimension of the original expanded surface feature of each node as the column;

[0191] The adjacency matrix, the two-dimensional feature matrix and the relationship feature matrix are input into the graph neural network to generate the target face expansion features of the person.

[0192] The device for constructing the social security coverage expansion feature provided in the embodiment of the present invention can execute the method for constructing the social security coverage expansion feature provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the method for constructing the social security coverage expansion feature.

[0193] See also Figure 4 , shows a schematic diagram of the structure of a screening device for social security coverage expansion personnel provided by an embodiment of the present invention. Figure 4 As shown, the device comprises:

[0194] The knowledge graph construction module 401 is used to construct a knowledge graph based on the source data of each person participating in social insurance;

[0195] A new structure screening module 402 is added, which is used to screen the graph substructure where the newly added personnel are located from the knowledge graph;

[0196] The original surface expansion feature generation module 403 is used to generate the original surface expansion feature for each node in the graph substructure;

[0197] A target face-expanding feature aggregation module 404 is used to aggregate the original face-expanding features of the same graph substructure into the target face-expanding features of the person;

[0198] A social security coverage expansion feature loading module 405 is used to load a plurality of social security coverage expansion features constructed according to a construction method of social security coverage expansion features;

[0199] The expanded personnel determination module 406 is used to determine the newly added personnel as potential personnel to participate in or upgrade social insurance when the target expansion feature successfully matches any of the social insurance expansion features.

[0200] In one embodiment of the present invention, the knowledge graph construction module 401 includes:

[0201] An entity screening module is used to screen entities related to social insurance participation in the source data of each person participating in social insurance;

[0202] A relationship query module, used to query the relationship between the entity representing the person and other entities in participating in social insurance in the source data of each person participating in social insurance;

[0203] A parameter query module, used to query the parameters of each entity in participating in social insurance in the source data of each person participating in social insurance;

[0204] The knowledge graph generation module is used to define the entities as nodes, the parameters as attributes of the nodes, and the relationships as edges, so as to construct a knowledge graph representing the characteristics of participating in social insurance.

[0205] In one embodiment of the present invention, the source data includes at least one of the following:

[0206] The basic data of the personnel, the insurance data of the personnel, the employment data of the personnel, and the basic data of the enterprise;

[0207] The relationship between the entity representing a person and the entity representing an enterprise is an employment relationship, the relationship between the entity representing a person and the entity representing insurance records is an insurance relationship, the relationship between the entity representing a person and the entity representing an occupation is an engagement relationship, the relationship between the entity representing a person and the entity representing a region is a residence relationship and / or an employment location relationship, and the relationship between the entity representing a person and the entity representing a family relationship is a kinship relationship;

[0208] The parameters of the personnel include at least one of name, gender, age, ID number, mobile phone number, and occupation; the parameters of the enterprise include at least one of enterprise name, enterprise type, address, contact number, and number of insured persons; the parameters of the insurance record include at least one of the insurance type, insurance time, and insurance payment details; the parameters of the occupation include at least one of the occupation name and occupation salary; the parameters of the region include address information of each level; and the parameters of the family relationship include relationship type.

[0209] In one embodiment of the present invention, the newly added structure screening module 402 includes:

[0210] A central node query module is used to query the node where the newly added person is located in the knowledge graph as the central node;

[0211] A neighbor node search module, used to search for other nodes adjacent to the central node in the knowledge graph as neighbor nodes;

[0212] The graph sub-node extraction sub-module is used to extract the central node, the neighboring nodes, the attributes of the central node, the attributes of the neighboring nodes, and the relationship between the central node and the neighboring nodes to obtain a graph sub-structure.

[0213] In one embodiment of the present invention, the expansion rule includes at least one of the following:

[0214] Those who have participated in occupational injury insurance but have not participated in basic pension insurance for enterprise employees;

[0215] Participated in work-related injury insurance according to the project, but not participated in the basic pension insurance for enterprise employees;

[0216] Those who have registered for employment but have not participated in the enterprise employee pension insurance;

[0217] Suspension of participation in enterprise employee pension insurance;

[0218] The participation and payment level of the basic medical insurance for urban and rural residents is greater than or equal to the preset threshold, and they have not participated in the enterprise employee pension insurance;

[0219] Belong to individual business owners and have not participated in the basic pension insurance for enterprise employees.

[0220] In one embodiment of the present invention, the original surface expansion feature generation module 403 includes:

[0221] The translation model loading module is used to load the preset translation model;

[0222] A triple data extraction module, used to extract the entity corresponding to each node, the relationship corresponding to the edge connected to the node and the attribute corresponding to the node in the graph substructure as triple data;

[0223] The feature translation module is used to input the triplet data into the translation model to generate the original expanded surface features of each node.

[0224] In one embodiment of the present invention, the target surface expansion feature aggregation module 404 includes:

[0225] The graph neural network loading module is used to load the preset graph neural network;

[0226] An adjacency matrix conversion module, used for converting the graph substructure into an adjacency matrix;

[0227] A two-dimensional feature matrix construction module is used to construct a two-dimensional feature matrix in the same graph substructure, taking the original expanded surface features of each node as rows;

[0228] A relationship feature matrix construction module is used to construct a relationship feature matrix in the same graph substructure, with the relationship between each node as the row and the dimension of the original expanded surface feature of each node as the column;

[0229] The adjacency matrix, the two-dimensional feature matrix and the relationship feature matrix are input into the graph neural network to generate the target face expansion features of the person.

[0230] In one embodiment of the present invention, the expansion personnel determination module 406 includes:

[0231] A similarity calculation module, used to calculate the similarity between the target expansion feature and any of the social security expansion features;

[0232] The similarity determination module is used to determine that the newly added person is a potential person to participate in or upgrade social insurance if the similarity is greater than or equal to a preset threshold.

[0233] The device for screening personnel for social security coverage expansion provided in the embodiment of the present invention can execute the method for screening personnel for social security coverage expansion provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method for screening personnel for social security coverage expansion.

[0234] See also Figure 5 , shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, blade servers, mainframe computers, and other suitable computers. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0235] like Figure 5 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 to the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0236] A number of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0237] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The processor 11 executes the various methods and processes described above, such as the method for constructing the social security expansion features or the method for screening social security expansion personnel.

[0238] In some embodiments, the method for constructing social security coverage expansion features or the method for screening social security coverage expansion personnel may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the method for constructing social security coverage expansion features or the method for screening social security coverage expansion personnel described above may be executed. Alternatively, in other embodiments, the processor 11 may be configured to execute the method for constructing social security coverage expansion features or the method for screening social security coverage expansion personnel in any other appropriate manner (for example, by means of firmware).

[0239] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0240] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0241] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0242] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0243] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0244] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0245] An embodiment of the present invention also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements a method for constructing social security coverage expansion features or a method for screening social security coverage expansion personnel as provided in any embodiment of the present invention.

[0246] In the process of implementation, the computer program product can be written in one or more programming languages ​​or a combination thereof to perform the computer program code of the present invention, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, using an Internet service provider to connect through the Internet).

[0247] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0248] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing social security coverage expansion features, characterized in that: include: Build a knowledge graph based on the source data of each person's social insurance participation; Filtering the graph substructure where the persons meeting the expansion rule are located from the knowledge graph; The expansion rules are used to identify potential people who can participate in or upgrade social insurance; Generating original expanded surface features for each node in the graph substructure; Aggregating the original face-expanding features of the same graph substructure into target face-expanding features of the person; Under the same expansion rule, the target expansion features are spliced ​​into social security expansion features.

2. The method according to claim 1, characterized in that: The knowledge graph is constructed based on the source data of each person participating in social insurance, including: Filter entities related to social insurance participation in the source data of social insurance participation of each person; Inquiring the relationship between the entity representing the person and other entities in social insurance participation in the source data of each person participating in social insurance; Querying the parameters of each entity in participating in social insurance in the source data of each person participating in social insurance; The entities are defined as nodes, the parameters are defined as attributes of the nodes, and the relationships are defined as edges, so as to construct a knowledge graph representing the characteristics of participating in social insurance.

3. The method according to claim 2, characterized in that The source data includes at least one of the following: The basic data of the personnel, the insurance data of the personnel, the employment data of the personnel, and the basic data of the enterprise; The relationship between the entity representing a person and the entity representing an enterprise is an employment relationship, the relationship between the entity representing a person and the entity representing insurance records is an insurance relationship, the relationship between the entity representing a person and the entity representing an occupation is an engagement relationship, the relationship between the entity representing a person and the entity representing a region is a residence relationship and / or an employment location relationship, and the relationship between the entity representing a person and the entity representing a family relationship is a kinship relationship; The parameters of the personnel include at least one of name, gender, age, ID number, mobile phone number, and occupation; the parameters of the enterprise include at least one of enterprise name, enterprise type, address, contact number, and number of insured persons; the parameters of the insurance record include at least one of the insurance type, insurance time, and insurance payment details; the parameters of the occupation include at least one of the occupation name and occupation salary; the parameters of the region include address information of each level; and the parameters of the family relationship include relationship type.

4. The method according to claim 2, characterized in that: The step of selecting graph substructures that meet the expansion rule from the knowledge graph includes: In the knowledge graph, the node where the person who meets the preset expansion rule is located is searched as the central node; In the knowledge graph, searching for other nodes adjacent to the central node as neighboring nodes; The central node, the neighboring nodes, the attributes of the central node, the attributes of the neighboring nodes, and the relationship between the central node and the neighboring nodes are extracted to obtain a graph substructure.

5. The method according to claim 4, characterized in that The expansion rule includes at least one of the following: Those who have participated in occupational injury insurance but have not participated in basic pension insurance for enterprise employees; Participated in work-related injury insurance according to the project, but not participated in the basic pension insurance for enterprise employees; Those who have registered for employment but have not participated in the enterprise employee pension insurance; Suspension of participation in enterprise employee pension insurance; The participation and payment level of the basic medical insurance for urban and rural residents is greater than or equal to the preset threshold, and they have not participated in the enterprise employee pension insurance; Belong to individual business owners and have not participated in the basic pension insurance for enterprise employees.

6. The method according to any one of claims 1 to 5, characterized in that The generating of original expanded surface features for each node in the graph substructure includes: Load the preset translation model; Extracting entities corresponding to each node, relationships corresponding to edges connected to the nodes, and attributes corresponding to the nodes from the graph substructure as triple data; The triplet data is input into the translation model to generate original expanded features of each of the nodes.

7. The method according to any one of claims 1 to 5, characterized in that The step of aggregating the original face-expanding features of the same graph substructure into the target face-expanding features of the person includes: Load the preset graph neural network; Converting the graph substructure into an adjacency matrix; In the same graph substructure, a two-dimensional feature matrix is ​​constructed with the original expanded features of each node as rows; In the same graph substructure, a relationship feature matrix is ​​constructed with the relationship between the nodes as rows and the dimensions of the original expanded features of the nodes as columns; The adjacency matrix, the two-dimensional feature matrix and the relationship feature matrix are input into the graph neural network to generate the target face expansion features of the person.

8. A method for screening personnel for social security expansion, characterized in that: include: Build a knowledge graph based on the source data of each person's social insurance participation; Filter the graph substructure where the newly added person is located from the knowledge graph; Generating original expanded surface features for each node in the graph substructure; Aggregating the original face-expanding features of the same graph substructure into target face-expanding features of the person; Loading multiple social security coverage expansion features constructed according to the method described in any one of claims 1 to 7; When the target coverage expansion feature successfully matches any of the social security coverage expansion features, the newly added person is determined as a potential person to participate in or upgrade social insurance.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the method for constructing social security coverage expansion features as described in any one of claims 1 to 7 or the method for screening social security coverage expansion personnel as described in claim 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for constructing social security coverage expansion features as described in any one of claims 1 to 7 or the method for screening social security coverage expansion personnel as described in claim 8.

Images