Talent recommendation method and device

Through the spectrum clustering algorithm, talent tags and employee relationship maps are generated, combined with employment needs, and the shortest path is determined to recommend expert talents, solving the problem of poor talent recommendation results in the existing technology, and achieving rapid and effective recommendation of cross-institutional expert talents.

CN112116331BActive Publication Date: 2024-11-08CHINA CONSTRUCTION BANK
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202011042735.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-28
Publication Date
2024-11-08
Estimated Expiration
2040-09-28

AI Technical Summary

Technical Problem

The failure of the existing technology to effectively utilize the talent link relationship in talent recommendations has led to poor rapid recommendation of cross-institutional expert talents.

Method used

The spectral clustering algorithm discovered by the community is used to generate talent tags, and the shortest path is determined through the employee relationship map for talent recommendation, and match it with the employment needs and employee relationship map.

Benefits of technology

It realizes rapid and effective recommendation of cross-institutional expert talents, forms a more effective recommendation relationship through the employee relationship map, and solves the problem of poor talent assistance results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112116331B_ABST
    Figure CN112116331B_ABST
Patent Text Reader

Abstract

The embodiment of the present invention provides a talent recommendation method and device, the method comprising: generating talent labels according to the spectral clustering algorithm discovered by the community; performing data processing according to the spectral clustering algorithm discovered by the community, forming entity and attribute information of employee relationships according to the processed data, and finally generating an employee relationship map; receiving employment needs; and determining the recommended talents that match the employment needs according to the employment needs, the talent labels and the employee relationship map. The embodiment of the present invention introduces the employee relationship map and combines it with the talent label library to automatically calculate the shortest employee relationship path of each recommended expert talent, and forms a more effective recommendation relationship through effective recommendation by employees.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a talent recommendation method and device. Background Art

[0002] Large and medium-sized enterprises have gathered a large number of experts and talents. However, many branches may not have the talents they need. When faced with new businesses and new products, they often need to explore on their own or ask for support from their superiors. There is a lack of an effective way to quickly seek expert guidance within the enterprise.

[0003] Figure 1 It is a flow chart of the talent recommendation method in the prior art. Figure 1 As shown, the existing common solutions mainly include: performing database search based on the basic data of the talent database, displaying the search results, and communicating and coordinating through contact information or superiors; a further approach is to build talent tags through basic data, search based on the talent tags, and then communicate and coordinate through contact information or superiors.

[0004] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0005] The existing technology adopts fuzzy query or precise query based on basic information of talents or talent tags, and does not consider effective links of talents in talent search and positioning. Although the contact information of talents can be viewed within a certain scope of authority, abrupt contact may have a counterproductive effect. Summary of the invention

[0006] The embodiment of the present invention provides a talent recommendation method and device to achieve rapid and effective recommendation of cross-institutional expert talents.

[0007] To achieve the above objectives, in a first aspect, an embodiment of the present invention provides a talent recommendation method, which includes:

[0008] Generate talent labels based on the spectral clustering algorithm discovered by the community;

[0009] Data is processed based on the spectral clustering algorithm discovered by the community, and entity and attribute information of employee relationships are formed based on the processed data, and finally an employee relationship map is generated;

[0010] Receive employment requests;

[0011] According to the employment demand, the talent tags and the employee relationship map, talents to be recommended that match the employment demand are determined.

[0012] In a second aspect, an embodiment of the present invention provides a talent recommendation device, which includes:

[0013] A label generation module, which is used to generate talent labels based on the spectral clustering algorithm discovered by the community;

[0014] The graph generation module is used to process data according to the spectral clustering algorithm discovered by the community, form entity and attribute information of employee relationships based on the processed data, and finally generate an employee relationship graph;

[0015] Demand interface module, used to receive employment requirements;

[0016] The talent recommendation module is used to determine the recommended talents matching the employment demand according to the employment demand, the talent tags and the employee relationship map.

[0017] In a third aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the talent recommendation method as described above.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer device, comprising:

[0019] one or more processors;

[0020] A storage device for storing one or more programs;

[0021] When the one or more programs are executed by the one or more processors, the one or more processors are enabled to implement the talent recommendation method as described above.

[0022] The above technical solution has the following beneficial effects:

[0023] The embodiment of the present invention generates talent labels by using a spectral clustering algorithm discovered by the community; performs data processing according to the spectral clustering algorithm discovered by the community, forms entity and attribute information of employee relationships according to the processed data, and finally generates an employee relationship map; receives employment needs; determines the recommended talents that match the employment needs according to the employment needs, the talent labels, and the employee relationship map; thereby solving the problem of rapid and effective recommendation of cross-institutional expert talents. The embodiment of the present invention introduces an employee relationship map and combines it with a talent label library to automatically calculate the shortest employee relationship path of each recommended expert talent, and forms a more effective recommendation relationship through effective recommendation by employees. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative work.

[0025] Figure 1 It is a flow chart of a talent recommendation method in the prior art;

[0026] Figure 2 is a flow chart of a talent recommendation method according to an embodiment of the present invention;

[0027] Figure 3 is another flow chart of the talent recommendation method according to an embodiment of the present invention;

[0028] Figure 4 This is the process of calculating employee labels using the spectral clustering algorithm in the embodiment of the present invention;

[0029] Figure 5 is a functional block diagram of a talent recommendation device according to an embodiment of the present invention;

[0030] Figure 6 It is a functional block diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 are within the scope of protection of the present invention.

[0032] The following is an explanation of the terms:

[0033] Spectral Clustering Algorithm: Spectral Clustering (SC) is a clustering method based on graph theory - it divides a weighted undirected graph into two or more optimal subgraphs, making the subgraphs as similar as possible and the distances between subgraphs as far as possible, so as to achieve the common clustering purpose. For a detailed introduction to the algorithm, please refer to: https: / / blog.csdn.net / u011596455 / article / details / 53443848 .

[0034] Employee relationship map: A relationship map constructed based on information such as employees' work processes, social activities, life events, etc. Employee relationships are a key factor affecting employee behavior, attitudes, work efficiency, and execution capabilities.

[0035] Employee labels: an intuitive representation of employee characteristics, including basic labels, performance labels, ability and quality labels, behavior labels, etc.

[0036] ETL: Extraction-Transformation-Loading, data extraction, transformation and loading. ETL is responsible for extracting data from distributed and heterogeneous data sources such as relational data and flat data files to a temporary intermediate layer for cleaning, transformation, and integration, and finally loading it into a data warehouse or data mart, becoming the basis for online analytical processing and data mining. In this embodiment, ETL technology can be replaced by data warehouse technology.

[0037] ES: elasticsearch, search engine. Specifically, ES is an open source distributed search engine based on RESTful web interface and built on Apache Lucene. At the same time, ES is also a distributed document database, in which each field can be indexed, and the data of each field can be searched. It can be horizontally expanded to hundreds of servers to store and process PB-level data. ES can store, search and analyze large amounts of data in a very short time. It is usually used as the core engine in complex search scenarios. ES is born for high availability and scalability. On the one hand, system expansion can be completed by upgrading hardware, which is called vertical or upward expansion (Vertical Scale / Scaling Up); on the other hand, adding more servers to complete system expansion is called horizontal expansion or outward expansion (Horizontal Scale / Scaling Out).

[0038] CLOB: Character Large Object, a type of file storage used in the database. In the Oracle database, LOB (Large Object) type fields have a large capacity (can hold up to 4GB of data), and a table can have multiple fields of this type, which is very flexible and suitable for business areas with very large data volumes (such as images, archives, etc.). LOB types include: BLOB, CLOB, and NCLOB. BLOB is a binary large object, suitable for storing non-text byte stream data (such as programs, images, audio and video, etc.). CLOB is related to character sets and is suitable for storing text data (such as historical archives, large works, etc.). In the ORACLE database, pictures, files, music and other information are stored in BLOB fields, and the files are first converted to binary and then stored. Articles or longer texts are stored in CLOB, which greatly facilitates future query, update, storage and other operations. NCLOB is an internal fixed-length multi-byte character large object.

[0039] HanLP: Han Language Processing, Chinese language processing package. The word segmentation methods in Hanlp Chinese natural language processing include standard word segmentation, NLP word segmentation, index word segmentation, N-shortest path word segmentation, CRF word segmentation, and fast dictionary word segmentation. There are many Chinese word segmentation algorithms, such as NLPIR of the Institute of Computing Technology of the Chinese Academy of Sciences, LTP of Harbin Institute of Technology, THULAC of Tsinghua University, Stanford word segmenter, Hanlp word segmenter, jieba word segmentation, IKAnalyzer, etc.

[0040] neo4J: Neo4j is a high-performance NOSQL graph database that stores structured data on the network instead of in tables. It is an embedded, disk-based, fully transactional Java persistence engine that stores structured data on the network (called a graph from a mathematical perspective) instead of in tables. Neo4j can also be seen as a high-performance graph engine that has all the features of a mature database. There are currently five major mainstream graph databases: Neo4j, TigerGraph, Amazon Neptune, JanusGraph, and ArangoDB.

[0041] One of the purposes of the embodiments of the present invention is to solve the problem of rapid and effective recommendation of cross-institutional expert talents. Existing talent recommendation methods mainly perform talent queries based on talent basic data, and a further approach is to construct talent tags and perform retrieval based on talent tags. The main problem with this method is that even if the talent information is matched, the effect of seeking assistance from expert talents is often not ideal because there is no intermediary to make an effective link.

[0042] The embodiment of the present invention introduces an employee relationship map and combines it with a talent tag library to automatically calculate the shortest employee relationship path for each recommended expert talent, and forms a more effective recommendation relationship through effective employee referrals.

[0043] The embodiments of the present invention relate to generating talent labels and employee relationship maps in the field of artificial intelligence, combining the two technologies to achieve intelligent recommendation of expert talents.

[0044] The embodiments of the present invention form talent labels through clustering and classification processing and other technologies; and construct an employee relationship map through knowledge graph technology.

[0045] The embodiments of the present invention introduce employee relationships based on a general database or tag search, perform the shortest path and most effectively accessible talent recommendation based on the employee relationship map, and use intermediaries to perform effective recommendations, thereby effectively solving the problem of poor talent assistance due to the lack of intersection.

[0046] Figure 2 FIG. 1 is a flow chart of a talent recommendation method according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0047] S110: Generate talent labels based on the spectral clustering algorithm discovered by the community.

[0048] Spectral Clustering (SC) is a clustering method based on graph theory: it divides a weighted undirected graph into two or more optimal subgraphs, making the subgraphs as similar as possible and the distances between subgraphs as far as possible, so as to achieve the purpose of common clustering.

[0049] S120: Process the data according to the spectral clustering algorithm discovered by the community, form entity and attribute information of employee relationships according to the processed data, and finally generate an employee relationship map.

[0050] Entity in knowledge graph: refers to something that is distinguishable and exists independently. Such as a person, a commodity, an animal, etc. Everything in the world is made up of specific things, which refers to entities. Entities are the most basic elements in knowledge graphs, and different entities have different relationships.

[0051] Attributes of the knowledge graph: Attributes are the abstraction of the relationship between entities. Then we can build a layer of relationships based on the attributes, such as person (type) → singer (property) → music (type).

[0052] S130: Receive employment requirements.

[0053] This step mainly includes the Human Resources Department's needs for the appointment and removal of leading cadres and the use of professional and technical staff. For example, a department promotes a deputy general manager and formulates basic conditions and requirements for the competition for the deputy general manager. It is necessary to extract leading cadres or employees who meet the corresponding conditions from the enterprise talent pool and talent tags to make accurate talent matching recommendations.

[0054] S140: Based on employment needs, talent tags and employee relationship maps, determine the talents to be recommended that match the employment needs.

[0055] In some exemplary implementations, step S140 may specifically include the following steps:

[0056] S141: Calculate the shortest employee relationship path based on employment needs, talent tags and employee relationship map;

[0057] S142: Based on the shortest employee relationship path, determine the talents to be recommended that match the employment needs.

[0058] In some exemplary implementations, step S110 may specifically include the following steps:

[0059] S111: Setting label rules and label categories corresponding to talents, where the label categories include any number of the following: basic labels, performance labels, ability and quality labels, and behavior labels;

[0060] S112: Use ETL data extraction, transformation and loading technology to extract structured data of talents and form relational database data;

[0061] S113: Perform calculations according to label rules to form basic labels and performance labels for talents;

[0062] S114: extracting unstructured document data from the talent document library; for example: paper documents of photographed employee resumes, paper documents of employee basic file information, paper forms of employee performance appraisal results, etc.

[0063] S115: Use a distributed search engine combined with a plug-in to extract employees' unstructured document data to form character-type large object CLOB data information of a relational database; employees refer to all employees of the enterprise, and talents refer to employees who have certain skills, are labeled accordingly and enter the talent pool;

[0064] S116: extracting keywords and summary information from the CLOB data information; this step may adopt various existing keyword and summary extraction algorithms;

[0065] S117: Calculate the extracted keywords and summary information with the tag rules to form corresponding tag data.

[0066] In some exemplary implementations, step S120 specifically includes the following steps:

[0067] S121: Use ETL technology to extract and analyze data from the human resources data mart to form a database entity relationship network between employees;

[0068] S122: constructing relationship graph ontology information based on database entity relationship data, and using a graph database to construct an employee relationship graph;

[0069] S123: Graph traversal based on graph database provides a graph traversal service interface based on entity attributes of the employee relationship graph.

[0070] In some exemplary implementations, step S140 specifically includes the following steps:

[0071] S141: talent matching through tag search service interface;

[0072] S142: searching for the shortest relationship path between the employment demand and the matched retrieved talents through the employee relationship graph algorithm, and retrieving and determining the talent list corresponding to the shortest relationship path;

[0073] S143: Applying the spectral clustering algorithm, displaying the talent list corresponding to the shortest relationship path determined by the search according to the matching degree calculated by the spectral clustering, and displaying the employee relationship information of the shortest relationship path.

[0074] Figure 3 FIG. 1 is another flow chart of the talent recommendation method according to an embodiment of the present invention. Figure 2 As shown, the technical solution of the embodiment of the present invention includes the following processing steps:

[0075] S1: Set label rules and label classification for expert talents, including basic labels, performance labels, ability and quality labels, behavior labels and other classification information.

[0076] Basic label rules: Extraction from the resume of the human resources system includes but is not limited to employee age range (under 25 years old, 25-35, 35-45, 45-55, over 55 years old), employee education definition (under bachelor's degree, bachelor's degree, master's degree, master's degree or above), employee's place of origin (North China, East China, South China, etc.), employee position information (bank leader, head office department manager level, division level, section level, below science and technology), etc.;

[0077] Performance labels: excellent, good, competent, below competent;

[0078] Capability and quality labels: distinguish the business capabilities and qualities of different lines, such as corporate business line, personal business line, financial technology line, risk compliance line, financial accounting line, etc. For example: financial technology line, big data senior, big data expert, artificial intelligence senior, artificial intelligence expert, artificial intelligence senior, artificial intelligence intermediate and artificial intelligence junior, etc.;

[0079] Behavior labels: For employee behavior management information generated in their daily work, different business lines are distinguished. For example, frequent lateness and overtime work are obtained based on attendance data; procurement work behavior is obtained based on procurement data; whether employees have violated regulations is obtained based on daily risk compliance, etc.

[0080] Figure 4 This is the process of calculating employee labels using the spectral clustering algorithm in the embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0081] S210: Data initialization operation: This step aggregates relevant structured and unstructured human resource data into the human resource data mart.

[0082] S220: Data ETL and normalization.

[0083] S230: Clustering the data according to the spectral clustering algorithm.

[0084] S240: Perform multiple generation.

[0085] S250: Determine whether the parameter adjustment iteration has achieved the desired effect. If yes, execute step S260; if no, execute step S230. This step is performed after multiple parameter adjustments, and the result or effect meets the requirements or needs of the business personnel. For example, the talents obtained from a talent pool after algorithm calculation and parameter adjustment meet the business needs.

[0086] S260: Select the talent label that best meets business needs.

[0087] S2: Use ETL (Extraction-Transformation-Loading) technology to extract structured data such as the basic information, work history, position information, job information, professional expertise, project experience, and assessment and evaluation results of experts and talents to form label data of the relational database. By performing calculations based on preset label rules, the personal basic label data and performance label data of experts and talents are formed.

[0088] S3: Extract unstructured document data such as work summaries, papers, and evaluation comments from the talent's document library. Use ES combined with Word\Excel\Pdf and other plug-ins to extract data from unstructured documents and form CLOB data information in the relational database.

[0089] S4: Use ETL technology to extract and analyze data from employee-related data to form a database entity relationship network between employees.

[0090] S5: Construct the relationship graph ontology information, and use the knowledge graph (such as Neo4j) to construct a knowledge graph of employee relationships including nodes such as personnel, institutions, projects, tasks, schools, teams, meetings, processes, organizations, administrative relationships, party-labor-union relationships, training relationships, marketing team relationships, task relationships, project relationships, meeting relationships, work circle relationships, process relationships, email relationships, etc.

[0091] S6: Provide a unified external expert talent recommendation engine interface to receive employment needs from various institutions, use natural language word segmentation and other technologies to decompose the query needs, first match experts through the label retrieval service interface, then find the shortest relationship path between the employment demand personnel and the retrieved experts and talents through spectral clustering, and return it to the expert talent recommendation engine interface.

[0092] S7: Display the retrieved expert talent list according to the matching degree, and display the employee relationship information of the shortest relationship path.

[0093] The present invention combines tag retrieval with employee relationship graph technology to recommend the shortest path and most effective talents based on the employee relationship graph, which solves the feasibility problem of expert talent recommendation. The specific advantages include:

[0094] The embodiment of the present invention can classify expert talent information into effective tags, and can perform rapid retrieval based on tag classification, priority, and matching degree;

[0095] The embodiment of the present invention can construct employee relationships through an employee relationship graph and quickly locate the shortest path relationship based on graph retrieval technology.

[0096] Figure 5 is a functional block diagram of a talent recommendation device according to an embodiment of the present invention. Figure 5 As shown, the talent recommendation device 300 includes:

[0097] A label generation module 310, for generating talent labels according to a spectral clustering algorithm discovered by the community;

[0098] A graph generation module 320 is used to process data according to the spectral clustering algorithm discovered by the community, form entity and attribute information of employee relationships according to the processed data, and finally generate an employee relationship graph;

[0099] The demand interface module 330 is used to receive employment demand;

[0100] The talent recommendation module 340 is used to determine the talents to be recommended that match the employment needs based on the employment needs, talent tags and employee relationship map.

[0101] In some exemplary embodiments, the recommendation module 340 is specifically used to: calculate the shortest employee relationship path based on employment needs, talent tags and employee relationship maps; and determine the recommended talents that match the employment needs based on the shortest employee relationship path.

[0102] In some exemplary implementations, the tag generation module 310 may be specifically configured to:

[0103] Set label rules and label categories corresponding to talents. Label categories include any number of the following: basic labels, performance labels, ability and quality labels, and behavior labels;

[0104] Use ETL data extraction, conversion and loading technology to extract structured data of talents and form data of relational database;

[0105] Calculate according to label rules to form basic labels and performance labels for talents;

[0106] Extract unstructured document data from talent's document library;

[0107] Use a distributed search engine combined with a plug-in to extract employees' unstructured document data and form character-based large object CLOB data information in a relational database;

[0108] Extract keywords and summary information from CLOB data information;

[0109] The extracted keywords and summary information are calculated with the label rules to form corresponding label data.

[0110] In some exemplary embodiments, the graph generation module 320 may be specifically used to:

[0111] Use ETL technology to extract and analyze data from the human resources data mart to form a database entity relationship network between employees;

[0112] Construct relational graph ontology information based on database entity relationship data, and use graph database to construct employee relationship graph;

[0113] Graph traversal based on graph database provides a graph traversal service interface based on entity attributes of the employee relationship graph.

[0114] In some exemplary implementations, the talent recommendation module 340 is specifically configured to:

[0115] Talent matching is performed through the tag search service interface;

[0116] Using the spectral clustering algorithm to find the shortest relationship path between the employment needs and the retrieved talents, and retrieving and determining the talent list corresponding to the shortest relationship path;

[0117] Applying the spectral clustering algorithm, the talent list corresponding to the shortest relationship path determined by the search is displayed according to the matching degree calculated by the spectral clustering, and the employee relationship information of the shortest relationship path is displayed.

[0118] The embodiment of the present invention introduces an employee relationship map and combines it with a talent tag library to automatically calculate the shortest employee relationship path for each recommended expert talent, and forms a more effective recommendation relationship through effective employee referrals.

[0119] The embodiments of the present invention relate to generating talent labels and employee relationship maps in the field of artificial intelligence, combining the two technologies to achieve intelligent recommendation of expert talents.

[0120] The embodiments of the present invention introduce employee relationships based on a general database or tag search, perform the shortest path and most effectively accessible talent recommendation based on the employee relationship map, and use intermediaries to perform effective recommendations, thereby effectively solving the problem of poor talent assistance due to the lack of intersection.

[0121] The embodiment of the present invention further provides an electronic device, such as Figure 6 As shown, it includes one or more processors 401 , a communication interface 402 , a memory 403 and a communication bus 404 , wherein the processor 401 , the communication interface 402 , and the memory 403 communicate with each other via the communication bus 404 .

[0122] Memory 403, used for storing computer programs;

[0123] The processor 401 is used to implement each step of any one of the above-mentioned talent recommendation methods when executing the program stored in the memory 403.

[0124] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the above electronic device and other devices.

[0125] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0126] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0127] An embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, each step of the above-mentioned continuous integration automated testing method is implemented.

[0128] Those skilled in the art may also understand that the various illustrative logical blocks, units, and steps listed in the embodiments of the present invention may be implemented by electronic hardware, computer software, or a combination of the two. In order to clearly demonstrate the interchangeability of hardware and software, the various illustrative components, units, and steps described above have generally described their functions. Whether such functions are implemented by hardware or software depends on the specific application and the design requirements of the entire system. Those skilled in the art may use various methods to implement the described functions for each specific application, but such implementation should not be understood as exceeding the scope of protection of the embodiments of the present invention.

[0129] The various illustrative logic blocks or units described in the embodiments of the present invention can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array or other programmable logic device, a discrete gate or transistor logic, a discrete hardware component, or any combination of the above. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors combined with a digital signal processor core, or any other similar configuration.

[0130] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0131] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, and readable storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A talent recommendation method, characterized in that: include: Generate talent labels based on the spectral clustering algorithm discovered by the community; Data is processed based on the spectral clustering algorithm discovered by the community, and entity and attribute information of employee relationships are formed based on the processed data, and finally an employee relationship map is generated; Receive employment requests; Determine the recommended talent that matches the employment demand according to the employment demand, the talent tag and the employee relationship map; The step of determining the recommended talent that matches the employment demand according to the employment demand, the talent tag and the employee relationship map specifically includes: Calculate the shortest employee relationship path according to the employment demand, the talent label and the employee relationship map; Determine the recommended talent that matches the employment demand based on the shortest employee relationship path; The generation of talent labels based on the spectral clustering algorithm discovered by the community specifically includes: Setting label rules and label categories corresponding to talents, wherein the label categories include any number of the following: basic labels, performance labels, ability and quality labels, and behavior labels; Use ETL data extraction, conversion and loading technology to extract structured data of talents and form relational database data; Perform calculations according to the label rules to form basic labels and performance labels for talents; Extract unstructured document data from talent's document library; Use a distributed search engine combined with a plug-in to extract employees' unstructured document data and form character-based large object CLOB data information in a relational database; Extract keywords and summary information from the CLOB data information; Calculating the extracted keywords and summary information with the tag rules to form corresponding tag data; The spectral clustering algorithm discovered by the community integrates data to form entity and attribute graph database information of employee relationships, and finally generates an employee relationship graph, which specifically includes: Use ETL technology to extract and analyze data from the human resources data mart to form a database entity relationship network between employees; Construct relational graph ontology information based on database entity relationship data, and use graph database to construct employee relationship graph; Graph traversal based on graph database provides a graph traversal service interface based on entity attributes of the employee relationship graph.

2. The method according to claim 1, characterized in that The step of determining the recommended talent that matches the employment demand according to the employment demand, the talent tag and the employee relationship map specifically includes: Talent matching is performed through the tag search service interface; Searching for the shortest relationship path between the employment demand and the retrieved talents through the employee relationship graph, and retrieving and determining a talent list corresponding to the shortest relationship path; Applying the spectral clustering algorithm, the talent list corresponding to the shortest relationship path determined by the search is displayed according to the matching degree calculated by the spectral clustering, and the employee relationship information of the shortest relationship path is displayed.

3. A talent recommendation device, characterized in that: include: A label generation module, which is used to generate talent labels based on the spectral clustering algorithm discovered by the community; The graph generation module is used to process data according to the spectral clustering algorithm discovered by the community, form entity and attribute information of employee relationships based on the processed data, and finally generate an employee relationship graph; Demand interface module, used to receive employment requirements; A talent recommendation module, used to determine the recommended talent that matches the employment demand based on the employment demand, the talent label and the employee relationship map; The recommendation module is specifically used to calculate the shortest employee relationship path according to the employment demand, the talent label and the employee relationship map; and determine the recommended talent that matches the employment demand according to the shortest employee relationship path; The label generation module is specifically used for: Setting label rules and label categories corresponding to talents, wherein the label categories include any number of the following: basic labels, performance labels, ability and quality labels, and behavior labels; Use ETL data extraction, conversion and loading technology to extract structured data of talents and form data of relational database; Perform calculations according to the label rules to form basic labels and performance labels for talents; Extract unstructured document data from talent's document library; Use a distributed search engine combined with a plug-in to extract employees' unstructured document data and form character-based large object CLOB data information in a relational database; Extract keywords and summary information from the CLOB data information; calculate the extracted keywords and summary information with the tag rules to form corresponding tag data; The graph generation module is specifically used for: Use ETL technology to extract and analyze data from the human resources data mart to form a database entity relationship network between employees; Construct relational graph ontology information based on database entity relationship data, and use graph database to construct employee relationship graph; Graph traversal based on graph database provides a graph traversal service interface based on entity attributes of the employee relationship graph.

4. The device according to claim 3, characterized in that The talent recommendation module is specifically used for: Talent matching is performed through the tag search service interface; Using the spectral clustering algorithm to find the shortest relationship path between the employment demand and the retrieved talents, and retrieving and determining a talent list corresponding to the shortest relationship path; Applying the spectral clustering algorithm, the talent list corresponding to the shortest relationship path determined by the search is displayed according to the matching degree calculated by the spectral clustering, and the employee relationship information of the shortest relationship path is displayed.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the talent recommendation method as claimed in claim 1 or 2 is implemented.

6. A computer device, characterized in that: It includes: one or more processors; A storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the talent recommendation method as claimed in claim 1 or 2.