Talent data screening method and device based on account association, equipment and medium

Abstract

The invention relates to the technical field of data analysis, and discloses a talent data screening method and device based on account association, equipment and a medium, and the method comprises the steps: constructing a resume vector set according to a pre-obtained personal resume corresponding to each employee account in a target enterprise, and generating a skill complementation map of each employee in the target enterprise based on the difference degree of each resume vector in the resume vector set, and identifying a department skill gap of the target enterprise according to the employee skill complementation map. And utilizing the department skill gap and preset demand skill talent distribution data to analyze gap talent data of the target enterprise, and based on the gap talent data, performing resume screening on a pre-acquired delivery resume to obtain target talent data. The target talent data screening result matching degree of the talent data screening method is improved.

Description

Technical Field

[0001] This invention relates to the field of data analysis technology, and in particular to a method, apparatus, device, and medium for screening talent data based on account association. Background Technology

[0002] In the field of corporate human resource management and team building, accurately identifying departmental skill gaps and matching suitable talent is a key step in improving organizational effectiveness.

[0003] Traditional methods often rely on manual analysis of employee resumes, making it difficult to efficiently extract the correlation between skill keywords and implicit skills in project experience. For example, current technologies cannot automatically identify skill keywords in resumes using natural language processing, nor can they accurately map project experience text to specific skill tags. Furthermore, the lack of a mechanism to deduplicate and merge synonymous skill expressions leads to redundant skill data, making it difficult to construct a structured employee skill map. Simultaneously, current technologies do not quantify skill proficiency into calculable indicators, failing to achieve digital representation of employee skills through vector transformation. This results in the selection of target talent failing to simultaneously meet the team's immediate skill needs and long-term talent reserve requirements. Summary of the Invention

[0004] This invention provides a talent data screening method, apparatus, device, and medium based on account association, to solve the problem of low matching degree of target talent data screening results in existing talent data screening methods on the market.

[0005] Firstly, a talent data screening method based on account association is provided, including: Construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target company obtained in advance. Based on the employee skills complementarity map, the departmental skills gaps of the target company are identified; Analyze the talent gap data of the target enterprise using the skills gap in the aforementioned departments and the pre-defined distribution data of skilled personnel in demand. Based on the talent shortage data, the pre-acquired application resumes are screened to obtain target talent data.

[0006] Secondly, a talent data screening device based on account association is provided, comprising: The graph construction module is used to construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target enterprise that have been obtained in advance. The gap identification module is used to generate a skill complementarity map of each employee in the target company based on the difference between each resume vector in the resume vector set, and to identify the departmental skill gaps of the target company based on the employee skill complementarity map. The talent analysis module is used to analyze the talent gap data of the target enterprise by using the skills gap of the department and the preset demand skills talent distribution data; The talent screening module is used to screen pre-acquired application resumes based on the talent shortage data to obtain target talent data.

[0007] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aforementioned talent data screening method based on account association.

[0008] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the aforementioned talent data screening method based on account association.

[0009] The above-mentioned scheme, implemented by the account-linked talent data screening method, device, equipment, and medium, can construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target enterprise, identify the departmental skill gaps of the target enterprise based on the employee skill complementarity map, analyze the talent gap data of the target enterprise using the departmental skill gaps and the preset demand skill talent distribution data, and screen the pre-acquired submitted resumes based on the talent gap data to obtain target talent data, thereby improving the efficiency of talent data screening and the matching degree of target talent. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of an application environment for a talent data screening method based on account association in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a talent data filtering method based on account association in one embodiment of the present invention; Figure 3 This is a schematic diagram of a talent data screening device based on account association in one embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a computer device according to an embodiment of the present invention; Figure 5 This is another structural schematic diagram of a computer device according to one embodiment of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] The talent data filtering method based on account association provided in this invention can be applied to, for example... Figure 1 In this application environment, the client communicates with the server via a network. The server can construct a resume vector set based on the personal resumes corresponding to each employee account within the target company, using the pre-acquired resumes from the client. It then identifies the departmental skill gaps within the target company based on the employee skill complementarity graph, analyzes the talent gap data of the target company using the departmental skill gaps and pre-set demand talent distribution data, and filters the pre-acquired resumes based on the talent gap data to obtain target talent data. This improves the efficiency of talent data construction and the matching degree of target talent. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The invention will be described in detail below through specific embodiments.

[0014] Please see Figure 2 As shown, Figure 2 A flowchart illustrating a talent data filtering method based on account association provided in an embodiment of the present invention includes the following steps: S1. Construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target company obtained in advance.

[0015] In this embodiment of the invention, the step of constructing a resume vector set based on the personal resumes corresponding to each employee account within the target enterprise is as follows: All employee resumes are acquired; a skill identification set is obtained by identifying skill keywords; a skill association set is obtained by identifying project experience text and associating it with skill tags; the names of projects participated in are also identified; the first two are then merged to obtain a total skill set; after deduplication and merging, a final skill set is obtained; the proficiency description of each skill is identified to obtain a score; a structured skill description text set is constructed based on the final skill set and proficiency scores; a comprehensive description text set incorporating project names is also constructed; this is then transformed into a resume vector set using a bag-of-words model of a preset technical field; finally, the vector set is uploaded to the server, where the server establishes an employee skill complementarity graph based on account associations.

[0016] In this embodiment of the invention, the target enterprise refers to an enterprise with recruitment needs, and obtaining the personal resumes of all employees in the target enterprise refers to obtaining the personal resumes filled out by the employees who have already joined the company based on their own work experience.

[0017] In this embodiment of the invention, the employee skill complementarity graph is a knowledge graph that focuses on the skill relationships among enterprise employees. Its core is to present the skill complementarity between employees through structured data.

[0018] In this embodiment of the invention, the step of constructing a resume vector set based on the pre-acquired personal resumes corresponding to each employee account within the target enterprise includes: Obtain the personal resumes of all employees within the target company, identify the skill keywords in the personal resumes, and obtain a skill identification set; Identify project experience text in the individual's resume, and associate skill tags with the project experience text to obtain a skill association set; Identify the names of the projects participated in from the project experience text; The skill identification set and the skill association set are merged to obtain the total skill set; The total set of skills is deduplicated and merged to obtain the final set of skills; Identify the skill proficiency description of each skill in the final skill set in the individual's resume, and generate a proficiency score based on the skill proficiency description; Based on the final skill set and the proficiency score, a structured skill description text is constructed to obtain a structured skill description text set; The names of the participating projects are integrated into the structured skill description text set to obtain a comprehensive description text set; The comprehensive descriptive text set is transformed into a vector using the bag-of-words model in the predefined technical field, resulting in a resume vector set.

[0019] In detail, the identification of skill keywords in the personal resume to obtain a skill identification set can be achieved by using natural language processing (NLP) technology (such as keyword matching and named entity recognition) to scan fields such as "skills" and "professional abilities" in the resume.

[0020] In detail, the process of identifying project experience text in the personal resume and associating skill tags based on the project experience text can be achieved by analyzing verb phrases in the project description (such as "responsible for building microservice architecture" → associated with the skill "microservice") to obtain the project experience text, and by combining industry knowledge bases (such as "developing WeChat mini programs" → associated with "front-end development" and "WeChat ecosystem") to associate skill tags.

[0021] Specifically, identifying the participating project names in the project experience text can be done by extracting project names through regular expressions or text pattern matching (such as "e-commerce system reconstruction project" in "2023 XX Company e-commerce system reconstruction project").

[0022] In detail, the deduplication and merging process for the total set of skills refers to eliminating duplicate expressions of synonymous skills, such as merging "machine learning algorithm" and "machine learning model" into "machine learning".

[0023] In detail, identifying the skill proficiency description of each skill in the final skill set within the individual's resume, and generating a proficiency score based on the skill proficiency description, transforms the textual proficiency into a calculable numerical indicator. For example, "understanding" corresponds to a score of 1, "proficient" corresponds to a score of 3, and "master" corresponds to a score of 5.

[0024] In detail, the structured skill description text is constructed based on the final skill set and the proficiency score in order to convert the skill data into a unified format for easy subsequent processing, such as "Employee ID_001: Python(5), Data Analysis(4)".

[0025] In detail, the identification of skill keywords in the personal resume can be achieved using natural language processing technology, such as extracting "Python" and "TensorFlow" from "proficient in Python and TensorFlow framework".

[0026] In detail, the construction of structured skill description text based on the final skill set and the proficiency score can be in the format of "ID: XXXX; [Skill 1, Score: 3; Skill 2, Score: 1; Skill 3, Score: 5]".

[0027] In detail, the bag-of-words model of the aforementioned technical field is a text vectorization tool designed for specific technical fields (such as IT, fintech, manufacturing, etc.) to transform unstructured technical text (such as employee resumes, project descriptions) into computable numerical vectors. Its core idea is to capture the technical terminology features in the text through a domain-customized word set (dictionary) and frequency statistics methods.

[0028] In this embodiment of the invention, by obtaining the personal resumes of all employees within the target enterprise and converting all employee resumes into vectors to obtain a set of resume vectors, the efficiency of subsequently establishing an employee skill complementarity map can be improved.

[0029] In this embodiment of the invention, generating a skill complementarity map of each employee within the target company based on the difference between each resume vector in the resume vector set includes: Extract the skill description vector contained in each resume vector in the resume vector set; The cosine similarity between each personal resume is calculated based on the skill description vector, and the complement of the cosine similarity is calculated to obtain the difference between personal resumes. Extract the project participation vector contained in each resume vector in the resume vector set; Calculate the project correlation degree between individual resumes based on the project participation vectors; The differences and the correlations between the projects are weighted and fused to obtain a complementarity score between individual resumes; Construct an adjacency matrix based on the complementarity score; Based on the adjacency matrix, a skill complementarity graph of each employee within the target enterprise is generated.

[0030] In detail, the cosine similarity is a parameter that measures the degree of similarity between two vectors; In detail, the calculation of the project correlation degree between individual resumes based on the project participation vector can be obtained by calculating the ratio of the number of jointly participated projects to the total number of projects. For example, if employee A participates in projects 1 and 2, and employee B participates in projects 2 and 3, the correlation degree is 1 / 3.

[0031] In detail, the adjacency matrix is ​​an n*n matrix (where n is the number of employees), and any element in the matrix represents the complementarity score between two employees.

[0032] In detail, the skill complementarity graph of each employee in the target enterprise is generated based on the adjacency matrix. The graph can be constructed using Gephi or Neo4j, where the node size represents the employee's skill centrality score and the edge weight corresponds to the complementarity score.

[0033] Specifically, the weighted fusion of the difference degree and the project correlation degree can be performed by assigning a weight of 0.5 to each of the difference degree and the project correlation degree.

[0034] In this embodiment of the invention, by constructing a set of resume vectors based on the personal resumes corresponding to each employee account within the target enterprise obtained in advance, the efficiency and accuracy of identifying departmental skill gaps in the target enterprise can be improved.

[0035] S2. Identify the departmental skill gaps of the target company based on the employee skill complementarity map.

[0036] In this embodiment of the invention, the departmental skills gap refers to the skills lacking in a technical team.

[0037] In this embodiment of the invention, identifying the departmental skill gaps of the target enterprise based on the employee skill complementarity map involves determining the skill complementarity matrix between employees in each department based on pre-acquired departmental division data and the employee skill complementarity map; then obtaining the team skill requirement list for each department; next, combining the skill complementarity matrix and the resume vector set to calculate the coverage score of each department for each professional skill in the requirement list; finally, based on the coverage score and a preset coverage threshold, identifying the skill gaps existing in the team.

[0038] In this embodiment of the invention, identifying the departmental skill gaps of the target enterprise based on the employee skill complementarity map includes: Based on the pre-acquired departmental division data and the employee skill complementarity map, a skill complementarity matrix is ​​determined among employees in each department; Summarize the pre-defined skill requirements of each department to obtain a team skill requirement list; The coverage score of each department for each professional skill in the team's skill requirement list is calculated based on the skill complementarity matrix and the resume vector set. Based on the coverage score and the preset coverage threshold, the departmental skill gaps for each department are identified.

[0039] In detail, the step of determining the skill complementarity matrix between employees in each department based on the pre-acquired departmental division data and the employee skill complementarity map is to identify the complementarity scores between employees in individual departments in the employee skill complementarity map and establish a skill complementarity matrix based on the identified complementarity scores.

[0040] In detail, the team skill requirements list can be set by referring to the skill configurations of benchmark teams in the same industry.

[0041] In this embodiment of the invention, calculating the coverage score of each department for each professional skill in the team skill requirement list based on the skill complementarity matrix and the resume vector set includes: Based on the resume vector set, identify the proficiency of each employee in each department for each professional skill in the team skill requirements list; Calculate the sum of the non-zero elements in each row of the skill complementarity matrix to obtain the skill centrality score of each employee in each department; The core and non-core talents in each department are identified based on the skill centrality score and the preset centrality threshold. The first proficiency score is obtained by summing the proficiency scores of all non-core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The second proficiency score is obtained by summing the proficiency scores of all core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The first proficiency score and the second proficiency score are weighted and summed to obtain the total proficiency score for each professional skill; Multiply the total number of employees in each department by the preset maximum proficiency score to obtain the maximum proficiency of each department for each professional skill in the team skill requirement list; Calculate the ratio of the total proficiency score to the maximum proficiency score to obtain the coverage score of each department for each professional skill in the team skill requirement list.

[0042] In detail, the resume vector set contains the proficiency of employees' relevant professional skills, and the proficiency score can be one of four scores: 0, 1, 3, or 5.

[0043] In detail, all elements in a single row of the skill complementarity matrix represent the complementarity score of a single employee with respect to all other employees. The skill centrality score of a single employee can be obtained by summing the non-zero elements in a single row of the skill complementarity matrix.

[0044] Specifically, the centrality threshold can be set according to a preset employee number ratio, and the value of the top 20% of all skill centrality scores can be selected as the centrality threshold.

[0045] In detail, the weighted sum of the first proficiency score and the second proficiency score is performed by assigning a higher weight parameter to the second proficiency score. For example, the first proficiency score is assigned a weight of 0.3 and the second proficiency score is assigned a weight of 0.7.

[0046] Specifically, the preset maximum proficiency score can be 5.

[0047] S3. Analyze the talent gap data of the target enterprise using the skills gap of the department and the preset demand skills talent distribution data.

[0048] In this embodiment of the invention, the step of analyzing the talent gap data of the target enterprise using the departmental skills gap and the preset demand skills talent distribution data includes: A domain knowledge graph is constructed by associating the personal resumes of all employees of all companies belonging to a preset industry within a preset target area. Obtain publicly available recruitment data from all companies in the preset industries within the target area, and quantify the demand frequency of skilled personnel in the recruitment market based on the recruitment data; Based on the domain knowledge graph, obtain the proficiency scores of all employees in the target area for each professional skill in the preset industry skills list, and obtain the overall proficiency score set; Identify the number of samples with the highest proficiency score in the overall proficiency score set to obtain the number of people who have mastered each professional skill in the industry skills list; The supply density is obtained by calculating the ratio of the number of people who have mastered each professional skill in the industry skills list to the total number of employees in the target company based on the number of people who have mastered it. The distribution of skilled personnel in demand is determined based on the supply density and the demand frequency. Based on preset dynamic weights, the skill gaps in the departments and the distribution of the required skills are integrated to obtain the talent gap data.

[0049] In this embodiment of the invention, the construction of a domain knowledge graph by associating the personal resumes of all employees of all enterprises belonging to a preset industry within a preset target area can be achieved by obtaining the personal resumes of all employees of enterprises in a preset industry within the target area (which may be within a city) through cross-account association. Based on a preset list of industry skills, the proficiency of all employees in the target area with each professional skill in the list of industry skills is identified according to the personal resumes. A graph is then constructed based on the list of industry skills and the proficiency of all employees in the target area with each professional skill in the list of industry skills.

[0050] In this embodiment of the invention, the identification of the distribution of demand skills based on the domain knowledge graph is obtained by analyzing the supply and demand relationship of each professional skill among enterprises in the target region.

[0051] In detail, obtaining publicly available recruitment data from all companies in the preset industries within the target area can be achieved by obtaining recruitment information (such as job descriptions and skill requirements) from multiple accounts, such as recruitment platforms and enterprise HR systems, through APIs or federated learning.

[0052] In detail, the quantitative calculation of the demand frequency of skilled personnel in the job market based on the recruitment data can be obtained by analyzing the recruitment data of various companies and calculating the ratio of the number of positions containing the target professional skills to the total number of positions. For example, if a total of 1,000 positions are recruited and 100 positions require the mastery of the target skill, then the demand frequency of the target skill is 0.1.

[0053] Specifically, determining the distribution of demanded skilled personnel based on the supply density and the demand frequency includes: Calculate the ratio of the demand frequency to the supply density to obtain the supply-demand ratio; The supply-demand ratio of each occupational skill in the industry skills list is calculated by multiplying it by the preset industry coverage to obtain the gap index of each occupational skill in the industry skills list. The required skills are identified based on the gap index and a preset gap index threshold; Based on the overall proficiency score set, employees with the highest proficiency in the required skills are identified, thus obtaining information on scarce talent. The university of the scarce talent is identified based on the scarce talent information and the recruitment data. By summarizing the required skills and the universities from which the scarce talents graduated, the distribution of talents with the required skills can be obtained.

[0054] In detail, the overall proficiency score set includes each employee's proficiency in each professional skill, as well as each employee's identity ID.

[0055] In detail, the step of identifying the required skills based on the gap index and a preset gap index threshold means determining the occupational skills that are greater than the gap index threshold as required skills.

[0056] In this embodiment of the invention, the analysis of the target company's talent shortage data using the departmental skills gap and the preset distribution data of required skilled personnel can be based on dynamic weights set according to specific recruitment needs. For example, when there is a need to recruit personnel who can promptly fill departmental skills gaps, the weight of the departmental skills gap can be set to 0.7, and the weight of the distribution data of required skilled personnel can be set to 0.3.

[0057] In this embodiment of the invention, to make the generated talent shortage data more accurate, before fusing the departmental skill gaps and the distribution of demanded skill talents based on preset dynamic weights to obtain the talent shortage data, the method further includes: Obtain internal recruitment information from the target company regarding the professional skills included in the skills gaps in the department; Based on the recruitment information, identify the number of positions needed for the specific skills required to fill the skills gaps in the department. Based on the employee skills complementarity map, identify the current number of employees in the target company who possess the professional skills included in the departmental skills gaps; Calculate the ratio of the number of positions to be filled to the number of existing positions to obtain the recruitment demand level; Obtain external recruitment information from all companies in the target industry within the target area; Calculate the recruitment demand growth rate for the required skills based on the external recruitment information; The recruitment demand level is added to the preset standard weight to obtain the skill gap weight value; The recruitment demand growth rate is added to the standard weight to obtain the demand skill weight value; The gap skill weight value and the demand skill weight value are normalized and summed to obtain the gap skill dynamic weight and the demand skill dynamic weight. The dynamic weights are obtained by summing the dynamic weights of the skills in short supply and the dynamic weights of the skills in demand.

[0058] In detail, obtaining the target company's internal recruitment information for the professional skills included in the department's skills gap can be done through the company's HR system or recruitment management platform to obtain recruitment information (such as job descriptions and the number of positions required) related to the professional skills included in the department's skills gap.

[0059] In detail, the step of identifying the number of job openings for the professional skills included in the skill gaps in the department based on the recruitment information can be achieved by extracting the number of job openings corresponding to each skill gap from internal recruitment information and summing them up to obtain the total number of job openings required.

[0060] In detail, identifying the number of existing employees in the target company who possess the professional skills included in the skill gaps of the department based on the employee skill complementarity map can be done by retrieving the number of all employees who have marked the skill gaps and whose proficiency scores are greater than 0, based on the employee skill complementarity map.

[0061] In detail, obtaining external recruitment information of all preset industry enterprises in the target area can be achieved by obtaining recruitment information of all preset industry enterprises in the target area (such as a city) through cross-account association (such as API, federated learning), and filtering out job data related to the required skills.

[0062] In detail, calculating the recruitment demand growth rate of the required skills based on the external recruitment information can refer to subtracting the recruitment demand from the number of people currently needed for recruitment in the same period last year from the number of people needed for recruitment in the external recruitment information, and then dividing the difference by the number of people needed for recruitment in the same period last year to obtain the recruitment demand growth rate.

[0063] Specifically, the standard weight can be 0.5.

[0064] In detail, the normalized summation of the gap skill weight value and the demand skill weight value can be achieved by summing the gap skill weight value and the demand skill weight value to obtain the total weight, and then dividing the gap skill weight value and the demand skill weight value by the total weight respectively.

[0065] S4. Based on the talent shortage data, the pre-acquired application resumes are screened to obtain target talent data.

[0066] In this embodiment of the invention, the process of screening pre-acquired application resumes based on the talent shortage data refers to screening the received application resumes based on the target talent data, and selecting resumes that match the target talent data.

[0067] In this embodiment of the invention, the step of filtering pre-acquired application resumes based on the talent shortage data to obtain target talent data includes: Extract the professional skills descriptions and graduating institutions contained in the submitted resumes to obtain a set of submission information; Calculate the semantic similarity between each delivery information in the delivery information set and the target talent data; Based on the semantic similarity and the preset similarity threshold, the submitted resumes are filtered to obtain target talent data.

[0068] In detail, the process of extracting the professional skill descriptions and graduating institutions contained in the application resume set to obtain the application information set can be achieved by extracting skill keywords and graduating institution keywords through the Named Entity Recognition (NER) model.

[0069] In detail, calculating the semantic similarity between each delivery information in the delivery information set and the target talent data can be done by calculating cosine similarity.

[0070] As can be seen, the above solution constructs a skills complementarity graph based on the resumes of employees in the target company. NLP technology is used to extract skill keywords and project experiences from the resumes, associate skill tags, and remove duplicates to form the final skills set. Skill proficiency is converted into numerical scores, and structured descriptive text is generated by combining project names. This is then transformed into resume vectors using a domain bag-of-words model. After uploading to the server, skill differences and project relevance are calculated using cosine similarity, and a weighted fusion is performed to obtain a complementarity score, constructing a visual graph. Laplace noise is added during the process to protect data privacy. Next, departmental skills gaps are identified based on the graph. A skills complementarity matrix is ​​determined based on departmental divisions, and a team skills requirement list is set by referring to industry benchmarks. The team's coverage score for the required skills is calculated and compared with a preset threshold to identify gaps. The coverage score needs to distinguish between core and non-core talent, and the proficiency scores of the two types of personnel are fused according to weights and the ratio is calculated. Finally, gap talent data is generated using gap and scarce talent distribution data. By constructing a domain knowledge graph through cross-account associations, the supply and demand relationship of industry skills within the target region is analyzed. The supply-demand ratio is obtained by calculating the ratio of demand frequency to supply density. Combined with industry coverage, a gap index is generated to identify in-demand skills and corresponding talent. Simultaneously, weights are dynamically adjusted based on internal and external recruitment data, integrating team gap and scarcity distribution data. Finally, resumes are screened based on target talent data. Skill descriptions and university information are extracted from resumes, and semantic similarity with target talent data is calculated. Matching talent is then selected based on thresholds.

[0071] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0072] In one embodiment, a talent data filtering device based on account association is provided, which corresponds one-to-one with the talent data filtering method based on account association described in the above embodiments. For example... Figure 3 As shown, the talent data screening device based on account association includes a graph construction module 101, a talent gap identification module 102, a talent analysis module 103, and a talent screening module 104. Detailed descriptions of each functional module are as follows: The graph construction module 101 is used to construct a set of resume vectors based on the personal resumes corresponding to each employee account in the target enterprise that have been obtained in advance. The gap identification module 102 is used to generate a skill complementarity map of each employee in the target company based on the difference degree of each resume vector in the resume vector set, and to identify the departmental skill gaps of the target company based on the employee skill complementarity map. Talent analysis module 103 is used to analyze the talent gap data of the target enterprise by using the department's skills gap and the preset demand skills talent distribution data; The talent screening module 104 is used to screen the pre-acquired application resumes based on the talent shortage data to obtain target talent data.

[0073] In one embodiment, when the graph construction module 101 performs the step of constructing a set of resume vectors based on the personal resumes corresponding to each employee account within the target enterprise, it is specifically used for: Obtain the personal resumes of all employees within the target company, identify the skill keywords in the personal resumes, and obtain a skill identification set; Identify project experience text in the individual's resume, and associate skill tags with the project experience text to obtain a skill association set; Identify the names of the projects participated in from the project experience text; The skill identification set and the skill association set are merged to obtain the total skill set; The total set of skills is deduplicated and merged to obtain the final set of skills; Identify the skill proficiency description of each skill in the final skill set in the individual's resume, and generate a proficiency score based on the skill proficiency description; Based on the final skill set and the proficiency score, a structured skill description text is constructed to obtain a structured skill description text set; The names of the participating projects are integrated into the structured skill description text set to obtain a comprehensive description text set; The comprehensive descriptive text set is transformed into a vector using the bag-of-words model in the predefined technical field, resulting in a resume vector set.

[0074] In one embodiment, the gap identification module 102, when performing the process of generating a skill complementarity map of each employee in the target enterprise based on the difference between each resume vector in the resume vector set, is specifically used for: Extract the skill description vector contained in each resume vector in the resume vector set; The cosine similarity between each personal resume is calculated based on the skill description vector, and the complement of the cosine similarity is calculated to obtain the difference between personal resumes. Extract the project participation vector contained in each resume vector in the resume vector set; Calculate the project correlation degree between individual resumes based on the project participation vectors; The differences and the correlations between the projects are weighted and fused to obtain a complementarity score between individual resumes; Construct an adjacency matrix based on the complementarity score; Based on the adjacency matrix, a skill complementarity graph of each employee within the target enterprise is generated.

[0075] In one embodiment, when the gap identification module 102 performs the step of identifying the departmental skill gaps of the target enterprise based on the employee skill complementarity map, it is specifically used for: Based on the pre-acquired departmental division data and the employee skill complementarity map, a skill complementarity matrix is ​​determined among employees in each department; Summarize the pre-defined skill requirements of each department to obtain a team skill requirement list; The coverage score of each department for each professional skill in the team's skill requirement list is calculated based on the skill complementarity matrix and the resume vector set. Based on the coverage score and the preset coverage threshold, the departmental skill gaps for each department are identified.

[0076] In one embodiment, the gap identification module 102, when performing the calculation of the coverage score of each department for each professional skill in the team skill requirement list based on the skill complementarity matrix and the resume vector set, is specifically used for: Based on the resume vector set, identify the proficiency of each employee in each department for each professional skill in the team skill requirements list; Calculate the sum of the non-zero elements in each row of the skill complementarity matrix to obtain the skill centrality score of each employee in each department; The core and non-core talents in each department are identified based on the skill centrality score and the preset centrality threshold. The first proficiency score is obtained by summing the proficiency scores of all non-core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The second proficiency score is obtained by summing the proficiency scores of all core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The first proficiency score and the second proficiency score are weighted and summed to obtain the total proficiency score for each professional skill; Multiply the total number of employees in each department by the preset maximum proficiency score to obtain the maximum proficiency of each department for each professional skill in the team skill requirement list; Calculate the ratio of the total proficiency score to the maximum proficiency score to obtain the coverage score of each department for each professional skill in the team skill requirement list.

[0077] In one embodiment, when the talent analysis module 103 performs the analysis of the target enterprise's talent gap data using the departmental skills gap and preset demand skills talent distribution data, it is specifically used for: Obtain publicly available recruitment data from all companies in the preset industries within the target area, and quantify the demand frequency of skilled personnel in the recruitment market based on the recruitment data; Based on the domain knowledge graph, obtain the proficiency scores of all employees in the target area for each professional skill in the preset industry skills list, and obtain the overall proficiency score set; Identify the number of samples with the highest proficiency score in the overall proficiency score set to obtain the number of people who have mastered each professional skill in the industry skills list; The supply density is obtained by calculating the ratio of the number of people who have mastered each professional skill in the industry skills list to the total number of employees in the target company based on the number of people who have mastered it. The distribution of demanded skilled personnel is determined based on the supply density and the demand frequency.

[0078] In one embodiment, the talent analysis module 103, when performing the step of determining the distribution of demanded skilled personnel based on the supply density and the demand frequency, is specifically used for: Calculate the ratio of the demand frequency to the supply density to obtain the supply-demand ratio; The supply-demand ratio of each occupational skill in the industry skills list is calculated by multiplying it by the preset industry coverage to obtain the gap index of each occupational skill in the industry skills list. The required skills are identified based on the gap index and a preset gap index threshold; Based on the overall proficiency score set, employees with the highest proficiency in the required skills are identified, thus obtaining information on scarce talent. The university of the scarce talent is identified based on the scarce talent information and the recruitment data. By summarizing the required skills and the universities from which the scarce talents graduated, the distribution of talents with the required skills can be obtained.

[0079] In one embodiment, when the talent analysis module 103 performs the analysis of the target enterprise's talent gap data using the departmental skills gap and preset demand skills talent distribution data, it is specifically used for: A domain knowledge graph is constructed by associating the personal resumes of all employees of all companies belonging to a preset industry within a preset target area. Obtain publicly available recruitment data from all companies in the preset industries within the target area, and quantify the demand frequency of skilled personnel in the recruitment market based on the recruitment data; Based on the domain knowledge graph, obtain the proficiency scores of all employees in the target area for each professional skill in the preset industry skills list, and obtain the overall proficiency score set; Identify the number of samples with the highest proficiency score in the overall proficiency score set to obtain the number of people who have mastered each professional skill in the industry skills list; The supply density is obtained by calculating the ratio of the number of people who have mastered each professional skill in the industry skills list to the total number of employees in the target company based on the number of people who have mastered it. The distribution of skilled personnel in demand is determined based on the supply density and the demand frequency. Based on a preset dynamic weight fusion of the departmental skill gaps and the distribution of demanded skilled personnel, gap talent data is obtained. In one embodiment, before performing the process of fusing the departmental skill gaps and the distribution of demanded skilled personnel based on preset dynamic weights to obtain gap talent data, the talent screening module 104 is further configured to: Obtain internal recruitment information from the target company regarding the professional skills included in the skills gaps in the department; Based on the recruitment information, identify the number of positions needed for the specific skills required to fill the skills gaps in the department. Based on the employee skills complementarity map, identify the current number of employees in the target company who possess the professional skills included in the departmental skills gaps; Calculate the ratio of the number of positions to be filled to the number of existing positions to obtain the recruitment demand level; Obtain external recruitment information from all companies in the target industry within the target area; Calculate the recruitment demand growth rate for the required skills based on the external recruitment information; The recruitment demand level is added to the preset standard weight to obtain the skill gap weight value; The recruitment demand growth rate is added to the standard weight to obtain the demand skill weight value; The gap skill weight value and the demand skill weight value are normalized and summed to obtain the gap skill dynamic weight and the demand skill dynamic weight. The dynamic weights are obtained by summing the dynamic weights of the skills in short supply and the dynamic weights of the skills in demand.

[0080] This invention provides a talent data screening device based on account association, constructing a skill complementarity graph based on the resumes of employees of target companies. It extracts skill keywords and project experiences from resumes using NLP technology, associates skill tags, removes duplicates, and merges them to form a final skill set. Skill proficiency is converted into numerical scores, and structured descriptive text is generated by combining project names. This text is then transformed into resume vectors using a domain bag-of-words model. After uploading to the server, skill differences and project relevance are calculated using cosine similarity, and a weighted fusion is performed to obtain a complementarity score, constructing a visual graph. Laplace noise is added during the process to protect data privacy. Next, the graph identifies departmental skill gaps. A skill complementarity matrix is ​​determined based on departmental divisions, and a team skill requirement list is set with reference to industry benchmarks. The team's coverage score for required skills is calculated and compared with a preset threshold to identify gaps. The coverage score needs to distinguish between core and non-core talent, and the proficiency scores of the two types of personnel are fused according to weights and the ratio is calculated. Finally, gap talent data is generated using gap and scarce talent distribution data. By constructing a domain knowledge graph through cross-account associations, the supply and demand relationship of industry skills within the target region is analyzed. The supply-demand ratio is obtained by calculating the ratio of demand frequency to supply density. Combined with industry coverage, a gap index is generated to identify in-demand skills and corresponding talent. Simultaneously, weights are dynamically adjusted based on internal and external recruitment data, integrating team gap and scarcity distribution data. Finally, resumes are screened based on target talent data. Skill descriptions and university information are extracted from resumes, and semantic similarity with target talent data is calculated. Matching talent is then selected based on thresholds.

[0081] Specific limitations regarding the account-linked talent data filtering device can be found in the limitations of the account-linked talent data filtering method described above, and will not be repeated here. Each module in the aforementioned account-linked talent data filtering device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0082] In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements server-side functions or steps of a talent data screening method based on account association.

[0083] In one embodiment, a computer device is provided, which may be a client, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements client-side functions or steps of a talent data screening method based on account association. In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: Construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target company obtained in advance. Based on the difference between each resume vector in the resume vector set, a skill complementarity map of each employee in the target company is generated. Based on the employee skills complementarity map, the departmental skills gaps of the target company are identified; Analyze the talent gap data of the target enterprise using the skills gap in the aforementioned departments and the pre-defined distribution data of skilled personnel in demand. Based on the talent shortage data, the pre-acquired application resumes are screened to obtain target talent data.

[0084] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: Construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target company obtained in advance. Based on the difference between each resume vector in the resume vector set, a skill complementarity map of each employee in the target company is generated. Based on the employee skills complementarity map, the departmental skills gaps of the target company are identified; Analyze the talent gap data of the target enterprise using the skills gap in the aforementioned departments and the pre-defined distribution data of skilled personnel in demand. Based on the talent shortage data, the pre-acquired application resumes are screened to obtain target talent data.

[0085] It should be noted that the functions or steps that can be implemented by the computer-readable storage medium or computer device described above can be referred to the relevant descriptions on the server side and client side in the foregoing method embodiments. To avoid repetition, they will not be described one by one here.

[0086] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0087] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0088] Finally, it should be noted that if any software tools or components not belonging to this company appear in the embodiments of the application, they are merely illustrative examples and do not represent actual use. The embodiments described above are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A talent data screening method based on account association, characterized in that, include: Construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target company obtained in advance. Based on the difference between each resume vector in the resume vector set, a skill complementarity map of each employee in the target company is generated. Based on the employee skills complementarity map, the departmental skills gaps of the target company are identified; Analyze the talent gap data of the target enterprise using the skills gap in the aforementioned departments and the pre-defined distribution data of skilled personnel in demand. Based on the talent shortage data, the pre-acquired application resumes are screened to obtain target talent data.

2. The talent data screening method based on account association as described in claim 1, characterized in that, The step of constructing a resume vector set based on the pre-acquired personal resumes corresponding to each employee account within the target enterprise includes: Obtain the personal resumes of all employees within the target company, identify the skill keywords in the personal resumes, and obtain a skill identification set; Identify project experience text in the individual's resume, and associate skill tags with the project experience text to obtain a skill association set; Identify the names of the projects participated in from the project experience text; The skill identification set and the skill association set are merged to obtain the total skill set; The total set of skills is deduplicated and merged to obtain the final set of skills; Identify the skill proficiency description of each skill in the final skill set in the individual's resume, and generate a proficiency score based on the skill proficiency description; Based on the final skill set and the proficiency score, a structured skill description text is constructed to obtain a structured skill description text set; The names of the participating projects are integrated into the structured skill description text set to obtain a comprehensive description text set; The comprehensive descriptive text set is transformed into a vector using the bag-of-words model in the predefined technical field, resulting in a resume vector set.

3. The talent data screening method based on account association as described in claim 1, characterized in that, The process of generating a skill complementarity map for each employee within the target company based on the difference between each resume vector in the resume vector set includes: Extract the skill description vector contained in each resume vector in the resume vector set; The cosine similarity between each personal resume is calculated based on the skill description vector, and the complement of the cosine similarity is calculated to obtain the difference between personal resumes. Extract the project participation vector contained in each resume vector in the resume vector set; Calculate the project correlation degree between individual resumes based on the project participation vectors; The differences and the correlations between the projects are weighted and fused to obtain a complementarity score between individual resumes; Construct an adjacency matrix based on the complementarity score; Based on the adjacency matrix, a skill complementarity graph of each employee within the target enterprise is generated.

4. The talent data screening method based on account association as described in claim 1, characterized in that, The step of identifying departmental skill gaps in the target company based on the employee skill complementarity map includes: Based on the pre-acquired departmental division data and the employee skill complementarity map, a skill complementarity matrix is ​​determined among employees in each department; Summarize the pre-defined skill requirements of each department to obtain a team skill requirement list; The coverage score of each department for each professional skill in the team's skill requirement list is calculated based on the skill complementarity matrix and the resume vector set. Based on the coverage score and the preset coverage threshold, the departmental skill gaps for each department are identified.

5. The talent data screening method based on account association as described in claim 4, characterized in that, The calculation of each department's coverage score for each professional skill in the team's skill requirement list based on the skill complementarity matrix and the resume vector set includes: Based on the resume vector set, identify the proficiency of each employee in each department for each professional skill in the team skill requirements list; Calculate the sum of the non-zero elements in each row of the skill complementarity matrix to obtain the skill centrality score of each employee in each department; The core and non-core talents in each department are identified based on the skill centrality score and the preset centrality threshold. The first proficiency score is obtained by summing the proficiency scores of all non-core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The second proficiency score is obtained by summing the proficiency scores of all core talents in each department for each professional skill in the team skill requirement list based on the proficiency of each professional skill. The first proficiency score and the second proficiency score are weighted and summed to obtain the total proficiency score for each professional skill; Multiply the total number of employees in each department by the preset maximum proficiency score to obtain the maximum proficiency of each department for each professional skill in the team skill requirement list; Calculate the ratio of the total proficiency score to the maximum proficiency score to obtain the coverage score of each department for each professional skill in the team skill requirement list.

6. The talent data screening method based on account association as described in claim 1, characterized in that, The analysis of the target company's talent shortage data using the departmental skills gap and the preset demand for skilled personnel distribution data includes: A domain knowledge graph is constructed by associating the personal resumes of all employees of all companies belonging to a preset industry within a preset target area. Obtain publicly available recruitment data from all companies in the preset industries within the target area, and quantify the demand frequency of skilled personnel in the recruitment market based on the recruitment data; Based on the domain knowledge graph, obtain the proficiency scores of all employees in the target area for each professional skill in the preset industry skills list, and obtain the overall proficiency score set; Identify the number of samples with the highest proficiency score in the overall proficiency score set to obtain the number of people who have mastered each professional skill in the industry skills list; The supply density is obtained by calculating the ratio of the number of people who have mastered each professional skill in the industry skills list to the total number of employees in the target company based on the number of people who have mastered it. The distribution of skilled personnel in demand is determined based on the supply density and the demand frequency. Based on preset dynamic weights, the skill gaps in the departments and the distribution of the required skills are integrated to obtain the talent gap data.

7. The talent data screening method based on account association as described in claim 6, characterized in that, Before obtaining the talent gap data by fusing the departmental skills gap and the distribution of required skills based on preset dynamic weights, the method further includes: Obtain internal recruitment information from the target company regarding the professional skills included in the skills gaps in the department; Based on the recruitment information, identify the number of positions needed for the specific skills required to fill the skills gaps in the department. Based on the employee skills complementarity map, identify the current number of employees in the target company who possess the professional skills included in the departmental skills gaps; Calculate the ratio of the number of positions to be filled to the number of existing positions to obtain the recruitment demand level; Obtain external recruitment information from all companies in the target industry within the target area; Calculate the recruitment demand growth rate for the required skills based on the external recruitment information; The recruitment demand level is added to the preset standard weight to obtain the skill gap weight value; The recruitment demand growth rate is added to the standard weight to obtain the demand skill weight value; The gap skill weight value and the demand skill weight value are normalized and summed to obtain the gap skill dynamic weight and the demand skill dynamic weight. The dynamic weights are obtained by summing the dynamic weights of the skills in short supply and the dynamic weights of the skills in demand.

8. A talent data screening device based on account association, characterized in that, include: The graph construction module is used to construct a set of resume vectors based on the personal resumes corresponding to each employee account within the target enterprise that have been obtained in advance. The gap identification module is used to generate a skill complementarity map of each employee in the target company based on the difference between each resume vector in the resume vector set, and to identify the departmental skill gaps of the target company based on the employee skill complementarity map. The talent analysis module is used to analyze the talent gap data of the target enterprise by using the skills gap of the department and the preset demand skills talent distribution data; The talent screening module is used to screen pre-acquired application resumes based on the talent shortage data to obtain target talent data.

9. A computer 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 computer program, it implements the talent data screening method based on account association as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the talent data screening method based on account association as described in any one of claims 1 to 7.

Images