Personnel information query method, device, electronic device and storage medium

Abstract

The embodiments of the present application provide a method, device, electronic device and storage medium for querying personnel information, and relate to the field of information query. The method comprises: receiving a first query request sent by a terminal, querying a knowledge graph according to a unique identifier of a person to be queried, determining an entity to be queried as a reference entity, and returning other entities directly associated with the reference entity as a first query result to the terminal, receiving a second query request sent by a terminal, and returning other entities directly associated with the entity to be queried using the knowledge graph according to the entity to be queried as a second query result to the terminal. The embodiments of the present application achieve fast and accurate retrieval of personnel information and other information associated with the person, reducing the time for retrieving information, avoiding more unnecessary retrieval processes, and saving resources at the same time.

Description

Technical Field

[0001] The present application relates to the field of information query technology. Specifically, the present application particularly relates to a method, device, electronic device and storage medium for querying personnel information. Background Art

[0002] With the development of the Internet era, more and more people are posting information on online platforms to find jobs. The amount of talent information in online recruitment is becoming larger and larger, and the talent level is becoming more and more complex. Customers often get lost in a large amount of talent information space and cannot successfully find the talents they need.

[0003] The traditional search method only has one level of interactive search, that is, the recruitment network platform provides an interactive search engine, and users enter requirements according to the columns provided above to conduct searches.

[0004] With this search method, the amount of search results obtained by customers is still very large, and they need to check the talent information one by one in the results to confirm whether it is the talent they need. This greatly limits the flexibility of the index, making it impossible to accurately and quickly search for the desired talent information, and the understanding of the talent information is not deep enough. Summary of the invention

[0005] Embodiments of the present invention provide a method, device, electronic device and storage medium for querying personnel information that overcome the above problems or at least partially solve the above problems.

[0006] In a first aspect, a method for querying personnel information is provided, the method comprising:

[0007] A first query request sent by a receiving terminal, wherein the first query request includes a unique identifier of a person to be queried;

[0008] According to the unique identifier of the person to be queried, a pre-constructed knowledge graph is queried, and an entity that records the unique identifier of the person to be queried in the knowledge graph is determined as a reference entity, and other entities directly associated with the reference entity are used as first target entities, and the first target entity is returned to the terminal as a first query result;

[0009] Receiving a second query request sent by the terminal, where the second query request includes a target entity to be queried determined by the terminal from the first query result;

[0010] Determine other entities directly associated with the target entity to be queried from the knowledge graph as second target entities, and return the second target entity to the terminal as a second query result.

[0011] Furthermore, the pre-built knowledge graph is queried according to the unique identifier of the person to be queried, and the steps of building the knowledge graph are also included, including:

[0012] Get sample resumes and extract information from the sample resumes;

[0013] Convert the information in the sample resume into entities, determine the identifiers of the entities, and determine the relationships between the entities based on the information in the sample resume;

[0014] Construct RDF triples based on the relationships between entities;

[0015] Import RDF triples into the graph database to build a knowledge graph.

[0016] Furthermore, constructing RDF triples according to the relationships between the entities also includes: performing deduplication processing on the entities according to the identifiers of the entities to obtain deduplication entities, specifically including:

[0017] Merge attributes of entities with the same identity;

[0018] For any two entities with inconsistent identifiers, determine whether the two entities are the same entity. If they are the same entity, merge the attributes of the two entities and retain the identifier of either entity as the identifier of the merged entity. Attributes are used to characterize the properties of entities and the relationship between entities.

[0019] Furthermore, the RDF triples are imported into the graph database to construct the knowledge graph, which also includes: removing duplicates from the RDF triples, specifically including:

[0020] If any two entities form two RDF triples according to the relationship between the two entities, the two RDF triples are merged.

[0021] Furthermore, the RDF triples are imported into the graph database to construct the knowledge graph, and then the indexing step is also included, including:

[0022] The receiving terminal sends the person information to be queried, converts the person information to be queried, obtains the conversion result of the person information to be queried, matches the person information to be queried with the entities in the database, obtains the entity corresponding to the person information to be queried, and uses the knowledge graph to obtain the knowledge graph query result;

[0023] The personnel information conversion results and the knowledge graph query results are integrated according to the preset structure to create an index.

[0024] Further, other entities directly associated with the target entity to be queried are determined from the knowledge graph as the second target entity, and the step of reprocessing the knowledge graph is also included, including:

[0025] In the knowledge graph, the target entity to be queried is used as the starting point, and other entities are traversed according to the preset random walk algorithm to obtain the vector representation of other traversed entities;

[0026] The vector representations of other entities are used according to the preset similarity algorithm to obtain the relationship between other entities and the target entity to be queried, and the relationship is written into the knowledge graph database.

[0027] In a second aspect, a device for querying personnel information is provided, comprising:

[0028] A first receiving module, configured to receive a first query request sent by a terminal, wherein the first query request includes a unique identifier of a person to be queried;

[0029] The first query module is used to query the pre-built knowledge graph according to the unique identifier of the person to be queried, determine the entity that records the unique identifier to be queried in the knowledge graph as the reference entity, take other entities directly associated with the reference entity as the first target entity, and return the first target entity to the terminal as the first query result;

[0030] A second receiving module, configured to receive a second query request sent by the terminal, wherein the second query request includes a target entity to be queried determined by the terminal from the first query result;

[0031] The second query module is used to determine other entities directly related to the target entity to be queried from the knowledge graph as the second target entity, and return the second target entity to the terminal as the second query result.

[0032] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method provided in the first aspect are implemented.

[0033] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method provided in the first aspect.

[0034] The personnel information query method, device, electronic device and storage medium provided in the embodiments of the present application realize the rapid and accurate retrieval of personnel information and other information associated with the personnel based on certain information of the personnel, thereby making personnel information query more efficient, reducing the time for retrieving information, avoiding more unnecessary retrieval processes, and saving resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0036] Figure 1 A network architecture diagram of a personnel information query system provided in an embodiment of the present application;

[0037] Figure 2 A flowchart of a method for querying personnel information provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of the process of constructing a knowledge graph provided in an embodiment of the present application;

[0039] Figure 4 The relationship between entities provided in the embodiment of the present application constitutes a structural display diagram;

[0040] Figure 5 A diagram showing the RDF triple structure for building a knowledge graph provided in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of the structure of a device for querying personnel information provided in an embodiment of the present application;

[0042] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present invention.

[0044] It will be understood by those skilled in the art that, unless expressly stated, the singular forms "one", "an" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0045] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0046] The present application provides a personnel information query method, device, electronic device and storage medium, which are intended to solve the above technical problems in the prior art.

[0047] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0048] First of all, the present application can be applied in various personnel information query scenarios, such as company employee information query, etc., wherein the personnel information query is to match the database information with the character strings such as the personnel's name, company name, position name, skill name, product name, school name, industry name, etc., and then obtain the detailed information of the personnel. The present application can query the information of the personnel through the knowledge graph based on a series of attribute data such as the personnel's name, company, school, industry, skills, position, etc. It can also query the information of the personnel through the knowledge graph based on the relationship data between the personnel and the company, school, skills, and position. Of course, there are other personnel query methods, which are not specifically limited in this application.

[0049] It should be understood that the personnel information query method provided in the present application can be applied to any computer device or system capable of personnel information query. Specifically, the personnel information query system can be run on Figure 1 In the network architecture shown, Figure 1 It is a network architecture diagram of a personnel information query system. The terminal 12 sends a query request to the server. The server 11 obtains information from the database according to the query request, and forms an information list and sends it to the terminal 12.

[0050] In order to more clearly illustrate the technical solution of the present application, the present application will be described in the personnel information query scenario in the following.

[0051] Figure 2 A flowchart of a method for querying personnel information provided in an embodiment of the present application, the embodiment of the present application at least includes the following steps:

[0052] S101. Receive a first query request sent by a terminal, where the first query request includes a unique identifier of a person to be queried.

[0053] In this embodiment, the first query request includes the ID of the person to be queried, the ID of the operator implementing the query, and the operation code, etc. The first query request must contain at least these three parts of information to realize the query. The operator's ID and operation code correspond to the query authority of different personnel information. The ID and operation code of the junior operator can only query the information of grassroots personnel, and the ID and operation code of the senior operator can query the information of grassroots personnel and internal personnel. If the junior operator privately queries the internal personnel information, it will be prompted that the query failed and the query is illegal, and the query interface will be directly exited. The specific situation depends on the actual scenario.

[0054] S102. Query a pre-constructed knowledge graph according to the unique identifier of the person to be queried, determine the entity that records the unique identifier of the person to be queried in the knowledge graph as a reference entity, take other entities directly associated with the reference entity as the first target entity, and return the first target entity to the terminal as the first query result.

[0055] The knowledge graph in this embodiment is pre-built. The knowledge graph is a graph-based data structure consisting of nodes and edges. Each node represents an entity, and each edge is the relationship between entities. The entity is the most basic element in the knowledge graph. It is a collection of certain things that are distinguishable and independently exist. Different entities have different relationships. Entities are generally nouns, representing abstract objects of people, things, and objects, such as employees, companies, schools, etc. The knowledge graph is composed of pieces of knowledge, and each piece of knowledge records the information of the person to be queried.

[0056] According to the ID of the person to be queried, other entities associated with the person to be queried are queried through the knowledge graph. For example, if the ID of the person to be queried is the name of the person, the entity with the unique identifier recorded in the knowledge graph is determined, that is, the personal information of the person to be queried. The personal information of the person to be queried is used as a reference entity to query other related entities. These related entities include the work experience, job title, graduation school, company, etc. of the person to be queried. These entities are called the first target entities and are sent back to the terminal as query results.

[0057] S103: Receive a second query request sent by the terminal, where the second query request includes a target entity to be queried determined by the terminal from the first query result.

[0058] It should be understood that the target entity to be queried is an entity in the first query result. For example, the first query result includes the work experience, job title, graduation school, company and other entities of the person to be queried, but these entities are not the final results of the query. It is also necessary to know the relevant information of the graduation school of the person to be queried, the size of the school, the comprehensive strength of the school, etc. Then the school is taken as the target entity to be queried, and the terminal sends a second query request, that is, requesting to query other entities associated with the school, and using the knowledge graph for a secondary query to obtain the query results.

[0059] S104. Determine other entities directly associated with the target entity to be queried from the knowledge graph as second target entities, and return the second target entity to the terminal as a second query result.

[0060] Specifically, an entity is selected from the first query result as the target entity to be queried, and other entities associated with the target entity are queried again through the knowledge graph, and these entities are returned to the terminal as query results. For example: the entity queried for the second time is the graduation school of the person to be queried. The knowledge graph can be used to learn the school's educational quality, school size, and school majors. If other entities need to be queried, other entities are still used as target entities for the query, and the knowledge graph is queried. This step is followed until the target information is obtained.

[0061] The personnel information query method, device, electronic device and storage medium provided in the embodiments of the present application realize the rapid and accurate retrieval of personnel information and other information associated with the personnel based on certain information of the personnel, thereby making personnel information query more efficient, reducing the time for retrieving information, avoiding more unnecessary retrieval processes, and saving resources.

[0062] Based on the above embodiment, as an optional embodiment, Figure 3 This is a schematic diagram of the process of constructing a knowledge graph in an embodiment of the present application. The pre-constructed knowledge graph is queried according to the unique identifier of the person to be queried, and the steps of constructing the knowledge graph are also included, including:

[0063] S201. Obtain a sample resume and extract information from the sample resume.

[0064] Obtain sample resumes from the Internet and extract all the information in the sample resumes through resume parsing technology, such as basic information, educational background, work experience, and all other information in the sample resumes.

[0065] S202: Convert the information in the sample resume into entities, determine the identifiers of the entities, and determine the relationships between the entities based on the information in the sample resume.

[0066] Table 1 is an entity conversion table. As shown in Table 1, all the extracted information is converted into a series of entities. These entities include positions, schools, project experience, personal information, work experience, companies, etc. Positions include position names, schools include school names, project experience includes skills, work items, companies include company names, company sizes, company locations, etc. It is also necessary to determine the relationship between each entity based on the information in the resume. Table 2 is a relationship conversion table. As shown in Table 2, for example, the relationship between entity Zhang San and entity school includes education experience, major, academic qualifications, study time, graduation time, etc. Figure 4 A knowledge graph structure diagram consisting of the relationships between entities.

[0067]

[0068]

[0069] Table 1: Entity conversion method table

[0070]

[0071]

[0072] Table 2: Relationship conversion table

[0073] S203: Construct RDF triples according to the relationships between entities.

[0074] Figure 5 The RDF triple structure diagram for constructing the knowledge graph is shown in the figure below. Figure 5 As shown, entities are organized into RDF triples in the form of (entity, relationship between entities, entity). RDF (Resource Description Framework) is essentially a data model that is used to describe entities. Simply put, it is a method and means of representing things, sometimes also called a statement. Statements constitute knowledge, and we also call the knowledge graph a piece of knowledge.

[0075] S204. Import the RDF triples into the graph database to construct a knowledge graph.

[0076] Import RDF triples into graph databases to build knowledge graphs. Graph databases include JanusGraph database, Neo4J database, etc. Neo4J database is a high-performance graph database with good stability. It is an embedded, disk-based Java persistence engine with complete transaction features, but it stores structured data on the network instead of in tables. Neo4j database can also be regarded as a high-performance graph engine with all the features of a mature database. This application mainly describes the technical points using Neo4J database as an example.

[0077] The embodiment of the present application extracts all the information in the resume, determines the relationship between entities based on the information in the resume, ensures that no information is omitted, and constructs it into RDF triples to facilitate the representation of the relationship between entities. The knowledge graph is constructed using a graph database to ensure the stability of the graph.

[0078] Based on the above embodiments, as an optional embodiment, RDF triples are constructed according to the relationship between entities, which also includes: deduplicating entities according to entity identifiers to obtain deduplicated entities, specifically including:

[0079] Merge attributes of entities with the same identity;

[0080] For any two entities with inconsistent identifiers, determine whether the two entities are the same entity. If they are the same entity, merge the attributes of the two entities and retain the identifier of either entity as the identifier of the merged entity. Attributes are used to characterize the properties of entities and the relationship between entities.

[0081] When obtaining sample resumes, this application may find that one resume uses multiple writing formats, but the information in the resumes is the same. This results in duplicate entities when all the information in the resumes is converted into entities. In order to ensure the uniqueness of each entity, it is necessary to eliminate duplicate entities, that is, after the resume information is converted into entities, each entity is assigned an identifier. The entity identifier is used to refer to the entity, which is equivalent to a symbol of the entity's identity. It should be understood that the same entity may be assigned multiple identifiers. For example, if someone retains two contact methods in his resume, when the entity identifier of the contact method is assigned, identifiers A and B are assigned respectively, but the attributes of the contact method entity are the same. There may also be multiple entities assigned the same label, such as For example, if a person attends the same university for both undergraduate and graduate studies, then when assigning entity identifiers, the identifier given to the school entity is C, but the attributes of the two school entities are different. The embodiment of the present application needs to perform duplicate processing on the entities according to the entity identifiers. First, any two entities with the same identifier are determined, and then the attributes of the two entities are merged, and only the unique entity and the identifier of one entity are retained. For example, if a person attends the same university for both undergraduate and graduate studies, then the identifier assigned to the school entity is C, but the attributes of the two entities are different. The two entities need to be merged into one entity, and the attributes of the two entities are superimposed and saved by the merged entity.

[0082] Secondly, for any two entities with different identifiers, an algorithm is used to deduplicate them. First, the contact information is used to deduplicate the person entity. For example, Zhang San uses the name Zhang San in a certain resume and his contact information is A. In another resume, he uses the name Li Si and the contact information is also A. Based on the contact information, it can be determined that Zhang San and Li Si are the same person. In this case, the attributes of Zhang San and Li Si are merged, and only one of the two entities, Zhang San or Li Si, is retained.

[0083] Secondly, the trajectory algorithm is used to remove duplicates. According to the trajectory algorithm, the relationship between the two entities and other entities is inferred. When the relationship between the two entities and other entities reaches and exceeds the preset similarity threshold, the two entities are determined to be the same entity, and the attributes of the two entities are merged. The trajectory algorithm is used for the person entity, that is, it defines what the person has accomplished in the spatial position in chronological order. This space-time sequence ((t1, s1, w1), (t2, s2, w2), ...) is called a trajectory, where t is time, s is a description of the spatial position, and w is a character representation of one of the titles, positions, projects, academic qualifications, and majors. The space-time sequence of each person constitutes a long area in the five-dimensional space-time. If the space-time areas of two people, that is, the intersection ratio of the trajectories, have a certain intersection ratio, then the two people are determined to be the same person. For example, Zhang San used different names on each resume when submitting his resume, but his educational experience and work experience are consistent. After the trajectory algorithm is calculated, the intersection of the trajectories of the two people exceeds the preset similarity threshold, then the two person entities can be determined to be the same entity.

[0084] You can also call the domain knowledge graph to make judgments and remove duplicates. If the domain knowledge graph results of any two entities are consistent, they are judged to be the same entity and the attributes of the two entities are merged. The domain knowledge graph is a description of a specific entity. For example, Zhang San studied at Wuhan University, where Wuhan University is a specific description of the school entity. If two entities are the same entity in the domain knowledge graph, they are merged.

[0085] Finally, an embedded representation method can be used to represent any two entities with vectors, and judge based on the similarity of the vector representation. If the similarity reaches and exceeds the preset threshold, the two entities are judged to be the same entity, and the attributes of the two entities are merged. Specifically, the graph embedding technology is used to convert the graph composed of entities into vectors. Graph embedding is a process of mapping graph data into low-density vectors. These vector relationships can generally only be represented by mathematics, statistics or specific subsets, but the embedded vector space has more flexible and rich calculation methods. Through calculation, if the similarity of the vectors corresponding to two entities is lower than a certain value, they are judged to be the same entity.

[0086] In order to identify whether any two entities are the same entity, the embodiment of the present application uses identification and algorithms for identification, and then superimposes the attributes of the same entity and removes duplicate entities, so that each entity is unique, ensuring that the relationship between entities is clearer, that is, improving the integrity of personnel information in the constructed knowledge graph.

[0087] Based on the above embodiments, as an optional embodiment, RDF triples are imported into a graph database to construct a knowledge graph, which also includes: deduplication of RDF triples, specifically including:

[0088] If any two entities form two RDF triples according to the relationship between the two entities, the two RDF triples are merged.

[0089] Specifically, the two entities in the corresponding positions of any two RDF triples are the same, and the relationship between the two entities is also the same, but the other attributes attached to the two relationships are different. It is necessary to merge the other attributes attached to the two relationships. For example, the RDF triple relationship formed by two entities A and B is P1 = (A-p1-B) and P2 = (A-p2-B), where A and B are entities referring to two people Zhang San and Li Si, and p1 and p2 are the relationship between the two entities, that is, Zhang San and Li Si are colleagues and p1 = p2 are both colleagues, but the attributes attached to the colleague relationship p1 include which specific company the colleague relationship is in. The attributes attached to the colleague relationship p2 include the duration of the colleague relationship and the projects they jointly worked on. When merging the attributes attached to the two relationships, if one of the attributes of p1 and p2 is an empty set and the other is a non-empty set, the non-empty set shall prevail. If both attributes are non-empty sets, both non-empty sets shall be retained and the same characters in the two non-empty sets shall be merged. For example: the attribute attached to p1 above is the colleague relationship in a certain company, and the attribute attached to p2 is that the two have been colleagues for a total of 5 years. Both attributes are not empty sets, so the same strings in them are retained and merged. If p1 and p2 do not have the same strings, both attributes are retained.

[0090] In the embodiment of the present application, for the same RDF triples, RDF triples of the same type of relationship are merged, and the accompanying attributes are grouped into one triple, so that the attributes attached to each entity are clearer, that is, the integrity of the personnel information in the constructed knowledge graph is ensured.

[0091] Based on the above embodiments, as an optional embodiment, RDF triples are imported into a graph database to construct a knowledge graph, and then the step of constructing an index is also included, including:

[0092] S301. Receive the person information to be queried sent by the terminal, convert the person information to be queried, obtain the conversion result of the person information to be queried, and match the person information to be queried with the entities in the database to obtain the entities corresponding to the person information to be queried, and use the knowledge graph to obtain the knowledge graph query result.

[0093] The receiving terminal sends the information of the person to be queried, which may be the original resume, certificate, etc. The information of the person to be queried is converted into character strings such as the person's name, contact information, work experience, education experience, skills, etc. through parsing service and conversion service, that is, according to the information in the resume and certificate.

[0094] According to the personnel information to be queried provided by the terminal, the personnel information to be queried can be character strings such as personnel name, company name, position name, skill name, school name, etc., which are associated and matched with the database entity ID to determine the entity ID of the information to be queried. For example: the terminal provides Zhang San's name, his graduation school is Wuhan University, and the name of the company he has worked for. By associating with the entity ID in the database, the entity ID of Zhang San and the entity ID of Wuhan University are obtained. According to the entity ID, the knowledge graph is used to query other related entities, or the personnel information to be queried provided by the terminal can be attribute information such as personnel, school, industry, skills, position, etc. For example: the terminal provides some attribute information such as the age of the personnel, what projects have been handled, and whether he has worked as an engineer in a certain company. The knowledge graph is used to query the entities associated with these attributes. The personnel information to be queried provided by the terminal can also be the relationship between the entities, such as being a colleague with Zhang San, studying at Wuhan University in 2006, and working in a certain company in 2010, etc. According to the relationship between the entities, the knowledge graph can be used to query the related entities.

[0095] S302: The personnel information conversion results and the knowledge graph query results are integrated according to a preset structure to create an index.

[0096] The strings converted into personnel names, personnel contact information, work experience, education experience, skills, etc. are associated with the entities queried using the knowledge graph, so that these strings are associated with the associated entities one-to-one. For example: the personnel name string corresponds to the personnel's basic information entity, and the education experience string corresponds to the personnel's education experience entity. In this way, an information list is constructed and written into the index.

[0097] The embodiment of the present application associates the knowledge graph query results and the personnel information character string to establish an index, which facilitates the subsequent viewing of the personnel information, that is, ensures that the personnel information is more organized.

[0098] Based on the above embodiments, as an optional embodiment, other entities directly associated with the target entity to be queried are determined from the knowledge graph as the second target entity, and the step of reprocessing the knowledge graph is also included, including:

[0099] S401. Taking the target entity to be queried as the starting point in the knowledge graph, traverse other entities according to a preset random walk algorithm to obtain vector representations of other traversed entities.

[0100] According to business needs, obtain the attributes of any target entity to be queried, and take the entity as the starting point, and perform depth or breadth traversal according to the random walk algorithm, in which the graph is regarded as a graph, the entity is regarded as a node, and the relationship between entities is regarded as an edge, that is, starting from a certain node, walk along the edge to check, and record all the traversed entities and the attributes of these entities. For example, Zhang San is the starting point, and the colleague relationship is the edge. You can walk to colleague A along the colleague relationship, and then walk to person B based on the friend relationship of colleague A. In this way, traverse and record the attributes of these entities, and then use the node vector representation method to calculate the vector representation of the traversed entity, and use these vector representations as the initial value of the next traversal to calculate the entity vector representation. For example: through the vector representation, the vector representation of the entity Zhang San is calculated as A, then the next traversal records another vector value B, which is calculated based on the current vector representation A of the entity Zhang San.

[0101] S402. Use the vector representations of other entities according to a preset similarity algorithm to infer the relationship between other entities and the target entity to be queried, and write the relationship into the knowledge graph database.

[0102] According to the vector value of the entity, the familiarity algorithm is used to infer the relationship between other entities and the target entity to be queried. Specific similarity algorithms include common node similarity algorithm, vector representation method, and graph convolutional neural network. For example, the entity Zhang San is taken as the starting point, and the colleague relationship is used to find the colleague Li Si entity, and then the person Wang Wu entity is found based on the friend relationship of the colleague Li Si entity. However, there is no clear record of the existence of a certain relationship between the two entities Zhang San and Wang Wu. Through the vectors of these two entities, the similarity is calculated, and it is inferred that there is an alumni relationship between the two entities Zhang San and Wang Wu, and this inferred alumni relationship is recorded in the knowledge graph.

[0103] Specifically, the reprocessing of the knowledge graph also includes other relationship processes indirectly obtained through existing relationship data, which is the reprocessing process of the graph. The overall process of this step is to regard the graph as a graph, the nodes are entities, the edges are relationships, and all relationship discoveries are based on the random walk judgment of the graph. The specific process is as follows: according to the needs of the business, a certain type of entity node is used as the starting point for depth or breadth traversal, and the traversal path of each node is recorded. In this way, we obtain a sequence composed of a series of entity relationship paths, analyze the obtained sequence, and use various algorithms to obtain a certain similarity of all nodes. These similarity analysis methods include the following: common node similarity, using The common nodes of any two nodes in the sequence (that is, there is a certain path between the node and the two nodes), and the ratio of the number of common nodes to the number of related paths of the two nodes is used; vector representation, using the transR and transE algorithms and the generated sequence, calculate the vector of each node, and then use the similarity of the vectors as the similarity of the two entity nodes; graph convolutional neural network, using graph convolutional neural network to directly analyze the similarity of any two entity nodes; according to business needs, the values ​​of various similarities are combined. When their values ​​are between a certain range of values, it is determined that the two entities have a given relationship.

[0104] The embodiment of the present application reprocesses the knowledge graph, takes the target entity as the starting point, and infers the unrecorded relationships with other entities through an algorithm, so that the personnel information is more comprehensive and the query is more extensive, that is, the personnel information recorded in the knowledge graph is ensured to be more accurate.

[0105] Figure 6 FIG. 1 is a schematic diagram of a device for querying personnel information according to an embodiment of the present application. Figure 6 As shown, the embodiment of the present application provides a device for querying personnel information, including:

[0106] The first receiving module 301 is used to receive a first query request sent by a terminal, wherein the first query request includes a unique identifier of a person to be queried;

[0107] The first query module 302 is used to query the pre-built knowledge graph according to the unique identifier of the person to be queried, determine the entity that records the unique identifier of the person to be queried in the knowledge graph as the reference entity, take other entities directly associated with the reference entity as the first target entity, and return the first target entity as the first query result to the terminal;

[0108] A second receiving module 303 is used to receive a second query request sent by the terminal, where the second query request includes a target entity to be queried determined by the terminal from the first query result;

[0109] The second query module 304 is used to determine other entities directly associated with the target entity to be queried from the knowledge graph as the second target entity, and return the second target entity to the terminal as the second query result.

[0110] The embodiment of the present application provides a device for querying personnel information, which specifically executes the above-mentioned method embodiment process. For details, please refer to the content of the embodiment of the above-mentioned method for querying personnel information, which will not be repeated here. The device for querying personnel information provided in the embodiment of the present application, through the first query request sent by the terminal, determines the entity of any unique identifier to be queried in the knowledge graph according to the unique identifier of the person to be queried, and sends other entities related to the entity back to the terminal as query results. The terminal can then send a second query request according to demand, and use the knowledge graph to query personnel information. The embodiment of the present application uses the knowledge graph to realize the fast and accurate retrieval of personnel information and other information associated with the person based on certain information of the person, so that the efficiency of personnel information query is higher, the time of retrieving information is reduced, more unnecessary retrieval processes are avoided, and resources are saved at the same time.

[0111] Furthermore, the first query module includes:

[0112] A resume acquisition module is used to obtain sample resumes and extract information from the sample resumes;

[0113] The entity conversion module is used to convert the information extracted from the sample resume into entities, and the information in the resume determines the relationship between entities;

[0114] The triple construction module is used to construct RDF triples based on the relationships between entities;

[0115] The graph construction module is used to import RDF triples into the graph database to build a knowledge graph.

[0116] Furthermore, the entity conversion module includes:

[0117] An entity first deduplication module, used to merge the attributes of entities with the same identifier;

[0118] The second entity deduplication module is used for any two entities with inconsistent identifiers to determine whether the two entities are the same entity. If they are the same entity, the attributes of the two entities are merged and the identifier of any one entity is retained as the identifier of the merged entity.

[0119] Furthermore, the triple building block includes:

[0120] The triple deduplication module is used for merging two RDF triples if any two entities form two RDF triples according to the relationship between the two entities.

[0121] Furthermore, the graph construction module includes:

[0122] The information receiving module is used to receive the information of the person to be queried sent by the terminal, convert the information of the person to be queried, obtain the conversion result of the information of the person to be queried, match the information of the person to be queried with the entities in the database, obtain the entity corresponding to the information of the person to be queried, and use the knowledge graph to obtain the knowledge graph query result;

[0123] The index building module is used to merge the personnel information conversion results with the knowledge graph query results according to the preset structure to create an index.

[0124] Furthermore, the second query module includes:

[0125] The entity traversal module is used to traverse other entities in the knowledge graph using the target entity to be queried as the starting point according to the preset random walk algorithm to obtain vector representations of other traversed entities;

[0126] The graph reprocessing module is used to convert the vector representations of other entities into a preset similarity algorithm, obtain the relationship between other entities and the target entity to be queried, and write the relationship into the knowledge graph database.

[0127] An electronic device is provided in an embodiment of the present application, and the electronic device includes: a memory and a processor; at least one program, which is stored in the memory for execution by the processor. Compared with the prior art, it can achieve: a first query request sent by a terminal, based on the unique identifier of the person to be queried, determining the entity of any unique identifier to be queried in the knowledge graph, and sending other entities related to the entity back to the terminal as query results. The terminal can then send a second query request as needed, and use the knowledge graph to query personnel information. The embodiment of the present application uses the knowledge graph to quickly and accurately retrieve personnel information and other information associated with the person based on certain information of the person, thereby making the personnel information query more efficient, reducing the time for retrieving information, avoiding more unnecessary retrieval processes, and saving resources at the same time.

[0128] In an alternative embodiment, an electronic device is provided, such as Figure 7 As shown, Figure 7 The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0129] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0130] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 7 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0131] The memory 4003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this.

[0132] The memory 4003 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the application code stored in the memory 4003 to implement the contents shown in the above method embodiment.

[0133] The embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when it is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment. The embodiment of the present application also provides a computer program product or a computer program, which includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the learning behavior monitoring method provided in various optional implementations of the above aspects. Compared with the prior art, the first query request sent by the terminal determines the entity of any unique identifier to be queried in the knowledge graph according to the unique identifier of the person to be queried, and sends other entities related to the entity back to the terminal as query results. The terminal can then send a second query request as needed, and use the knowledge graph to query personnel information. The embodiment of the present application uses the knowledge graph to obtain the basic information of the person and other information related to the basic information of the person based on certain information of the person, thereby making the query of personnel information more efficient. According to other information related to the person, the knowledge graph is used to query and obtain more related information, so that the obtained information is more extensive and more accurate, and the way of querying information is also more flexible.

[0134] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0135] The above are only some embodiments of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for querying personnel information, characterized in that: include: A first query request sent by a receiving terminal includes a unique identifier of a person to be queried, a unique identifier and an operation code of an operator who performs the query, and the unique identifier and the operation code of the operator correspond to the query authority; According to the unique identifier of the person to be queried, a pre-constructed knowledge graph is searched, and an entity that records the unique identifier of the person to be queried in the knowledge graph is determined as a reference entity, and other entities directly associated with the reference entity are used as first target entities, and the first target entity is returned to the terminal as a first query result; receiving a second query request sent by the terminal, wherein the second query request includes a target entity to be queried determined by the terminal from the first query result; Determine other entities directly associated with the target entity to be queried from the knowledge graph as second target entities, and return the second target entity to the terminal as a second query result; The step of determining other entities directly associated with the target entity to be queried from the knowledge graph as the second target entity also includes the step of reprocessing the knowledge graph: In the knowledge graph, taking the target entity to be queried as the starting point, traversing other entities according to a preset random walk algorithm, and obtaining vector representations of other traversed entities; Determine the similarity between the other entities and the target entity to be queried according to a preset similarity algorithm for the vector representation of the other entities. If the similarity is within a preset range, determine that there is a relationship between the other entities and the entity to be queried, and write the relationship into the knowledge graph database; The process of building a knowledge graph includes the following steps: removing duplicate entities and obtaining the following entities: For any two entities with inconsistent identifiers, the trajectory algorithm, calling the domain knowledge graph and the embedded expression method are used to determine whether the two entities are the same entity. If they are the same entity, the attributes of the two entities are merged, and the identifier of any one entity is retained as the identifier of the merged entity. The attributes are used to characterize the properties of the entity and the relationship between entities.

2. The method for querying personnel information according to claim 1, characterized in that: The step of querying the pre-built knowledge graph according to the unique identifier of the person to be queried also includes the step of constructing the knowledge graph: Obtain a sample resume and extract information from the sample resume; Convert the information in the sample resume into entities, determine the identifiers of the entities, and determine the relationships between the entities based on the information in the sample resume; Constructing RDF triples according to the relationships between the entities; Import the RDF triples into a graph database to construct a knowledge graph.

3. The method for querying personnel information according to claim 2, characterized in that: The step of constructing the RDF triples according to the relationships between the entities also includes: performing deduplication processing on the entities according to the identifiers of the entities to obtain deduplicated entities, specifically: Merges properties of entities with the same identity.

4. The method for querying personnel information according to claim 2, characterized in that: The step of importing the RDF triples into a graph database to construct a knowledge graph also includes: removing duplicates from the RDF triples, specifically: If any two entities form two RDF triples according to the relationship between the two entities, the two RDF triples are merged.

5. The method for querying personnel information according to claim 2, characterized in that: The step of importing the RDF triples into the graph database to construct a knowledge graph, and then constructing an index, includes: The information of the person to be queried sent by the receiving terminal is converted to obtain the conversion result of the information of the person to be queried, and the entity corresponding to the information of the person to be queried is obtained by matching the information of the person to be queried with the entity in the database, and the entity is obtained by using the knowledge graph to obtain the knowledge graph query result; The personnel information conversion result and the knowledge graph query result are integrated according to a preset structure to create an index.

6. A device for querying personnel information, characterized in that: include: A first receiving module, configured to receive a first query request sent by a terminal, wherein the first query request includes a unique identifier of a person to be queried, a unique identifier of an operator implementing the query, and an operation code, wherein the unique identifier of the operator and the operation code have a corresponding relationship with the query authority; A first query module is used to query a pre-constructed knowledge graph according to the unique identifier of the person to be queried, determine the entity in the knowledge graph that records the unique identifier of the person to be queried as a reference entity, use other entities directly associated with the reference entity as first target entities, and return the first target entity to the terminal as a first query result; A second receiving module, configured to receive a second query request sent by the terminal, wherein the second query request includes a target entity to be queried determined by the terminal from the first query result; A second query module, configured to determine other entities directly associated with the target entity to be queried from the knowledge graph as second target entities, and return the second target entity to the terminal as a second query result; The second query module determines other entities directly associated with the target entity to be queried from the knowledge graph as the second target entity, and the step of reprocessing the knowledge graph is also included: In the knowledge graph, taking the target entity to be queried as the starting point, traversing other entities according to a preset random walk algorithm, and obtaining vector representations of other traversed entities; Determine the similarity between the other entities and the target entity to be queried according to a preset similarity algorithm for the vector representation of the other entities. If the similarity is within a preset range, determine that there is a relationship between the other entities and the entity to be queried, and write the relationship into the knowledge graph database; The process of building a knowledge graph includes the following steps: removing duplicate entities and obtaining the following entities: For any two entities with inconsistent identifiers, the trajectory algorithm, calling the domain knowledge graph and the embedded expression method are used to determine whether the two entities are the same entity. If they are the same entity, the attributes of the two entities are merged, and the identifier of any one entity is retained as the identifier of the merged entity. The attributes are used to characterize the properties of the entity and the relationship between entities.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method for querying personnel information as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions enable the computer to execute the steps of the method for personnel information query as described in any one of claims 1 to 5.

Images