Intelligent identification method based on talent digital intelligent brain system

Abstract

The invention relates to the technical field of human resource management, in particular to an intelligent recognition method based on a talent digital intelligent brain system, which comprises the following steps: firstly, constructing a talent database based on the talent digital intelligent brain system; then, enterprise recruitment demand texts, historical recruitment data and industry trend information are collected, and key skills, experience requirements, cultural integrating degrees and other core indexes are extracted; combining enterprise historical data and industry dynamics, and identifying potential demands which are not clearly expressed by the enterprise through an association rule mining algorithm; integrating the dominant demand and the potential demand to form a structured enterprise recruitment demand set; candidates are intelligently screened based on a talent digital intelligent brain system in combination with an enterprise recruitment demand set, after screening is completed, the system generates a personalized recommendation report, and in this way, the problems that in current talent recruitment and management work, the recruitment efficiency is low, and the recruitment time is short are solved. And precious talent resources are wasted.

Description

Technical Field

[0001] The present invention relates to the technical field of human resource management, and in particular to an intelligent identification method based on a talent digital brain system. Background Art

[0002] With the rapid development of globalization and informatization, talent management plays an increasingly important role in national development and social progress. In the digital economy era, the digital and intelligent transformation of talent management has become an irreversible trend. The digitalization of talent management not only means leveraging digital technologies to improve the efficiency and quality of talent management, but also aims to optimize the allocation and efficient utilization of talent resources through advanced technologies such as big data and artificial intelligence.

[0003] However, in the current talent recruitment and management work, it is difficult to accurately match the actual needs of enterprises, resulting in low recruitment efficiency and waste of valuable talent resources. Summary of the Invention

[0004] The purpose of this invention is to provide an intelligent recognition method based on a talent digital brain system, aiming to solve the technical problems in the existing technology that it is difficult to accurately match the actual needs of enterprises in current talent recruitment and management work, resulting in low recruitment efficiency and waste of valuable talent resources.

[0005] To achieve the above objectives, the present invention adopts an intelligent identification method based on a talent digital brain system, comprising the following steps:

[0006] First, build a talent database based on the talent digital brain system;

[0007] Then collect the company's recruitment demand text, historical recruitment data and industry trend information, and pre-process the demand data;

[0008] Use natural language processing technology to analyze pre-processed demand data and extract core indicators such as key skills, experience requirements, and cultural fit;

[0009] Combining historical enterprise data with industry trends, we assign weights to various needs and use association rule mining algorithms to identify potential needs that are not clearly stated by the enterprise.

[0010] Integrate explicit demand and potential demand to form a structured set of corporate recruitment needs;

[0011] Based on the talent digital brain system and combined with the company's recruitment needs set, a multi-dimensional matching algorithm is used to intelligently screen candidates. After the screening is completed, the system generates a personalized recommendation report.

[0012] The specific method of building a talent database based on the talent digital brain system is as follows: the talent digital brain system builds a talent database by integrating multi-source heterogeneous data;

[0013] Specifically, a rule-based data cleaning method is used to remove duplicate, erroneous, and incomplete data records. At the same time, talent information is standardized, such as uniformly coding skill names and educational levels.

[0014] Use data deduplication technology based on hash algorithm to ensure the uniqueness of each record in the talent database;

[0015] The talent database constructed supports multi-dimensional retrieval and analysis of talent information.

[0016] Among them, BERT or Transformer models are used to parse preprocessed demand data to extract key skills, experience requirements and cultural fit.

[0017] The specific method of assigning weights to various requirements, combining historical enterprise data with industry dynamics, is as follows:

[0018] Combine the company's historical data with industry dynamics and dynamically adjust the weights of various requirements based on AHP or entropy weight method;

[0019] At the same time, the TF-IDF algorithm is used to evaluate the importance of industry trend keywords.

[0020] Among them, according to the feature type of the enterprise recruitment demand set, cosine similarity, Euclidean distance or Manhattan distance is selected to calculate the similarity with the talents in the talent database to obtain the candidate set;

[0021] According to the weight of each feature in the recruitment demand set, the similarity of candidates in the candidate set is weighted summed;

[0022] The comprehensive matching degree of the candidates is obtained by weighted summation, and the candidates are ranked based on the comprehensive matching degree. The candidates with higher matching degrees are ranked higher.

[0023] Among them, after the system completes the candidate screening and generates a personalized recommendation report:

[0024] Use the K-means clustering algorithm to group candidates; evaluate the matching degree through decision tree or random forest models; and generate a visual report containing talent profiles, skill tags, and matching scores.

[0025] Among them, feedback from companies and candidates is collected, categorized, and filed according to company, candidate, and feedback type;

[0026] Use data analysis tools and methods to deeply explore the reasons behind feedback data;

[0027] Based on the problem diagnosis results, the multi-dimensional matching algorithm is optimized in a targeted manner.

[0028] Among them, when collecting feedback from companies and candidates,

[0029] For enterprises: A dedicated feedback portal is set up in the Talent Digital Brain system, allowing enterprises to submit their evaluations of recommended candidates and adjust recruitment needs at any time;

[0030] For the talent side: After the candidates complete the interview, an evaluation link will be sent via SMS or email to collect their feedback on the interview process, the matching degree of the recommended position, and their own competitiveness assessment.

[0031] Collect candidates' comments and discussions on social media and recruitment platforms to understand their views and experiences with the positions recommended by the system.

[0032] Among them, as the talent database and corporate recruitment demand sets are continuously updated, the matching algorithm is rerun regularly to obtain the latest recommendation results.

[0033] The intelligent identification method based on the talent digital brain system of the present invention, when used specifically, first builds a talent database based on the talent digital brain system; then collects corporate recruitment demand texts, historical recruitment data and industry trend information, and pre-processes the demand data; uses natural language processing technology to parse the pre-processed demand data, and extracts core indicators such as key skills, experience requirements and cultural fit; combines corporate historical data with industry dynamics to assign weights to various demands, and uses association rule mining algorithms to identify potential demands that are not clearly stated by the company; integrates explicit demands with potential demands to form a structured corporate recruitment demand set; based on the talent digital brain system and combined with the corporate recruitment demand set, uses a multi-dimensional matching algorithm to intelligently screen candidates. After the screening is completed, the system generates a personalized recommendation report, thereby solving the technical problem in the current talent recruitment and management work that it is difficult to accurately match the actual needs of the company, resulting in low recruitment efficiency and a waste of precious talent resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. 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 any creative work.

[0035] Figure 1 It is a flow chart of the intelligent identification method based on the talent digital brain system of the present invention. DETAILED DESCRIPTION

[0036] The embodiments of the present invention are described in detail below. Examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, but should not be understood as limiting the present invention.

[0037] See also Figure 1 , Figure 1 It is a flow chart of the intelligent identification method based on the talent digital brain system of the present invention.

[0038] The present invention provides an intelligent identification method based on a talent digital brain system, comprising the following steps:

[0039] S1. First, build a talent database based on the talent digital brain system;

[0040] For this specific implementation, the Talent Digital Brain System builds a talent database by integrating multi-source heterogeneous data;

[0041] Specifically, a rule-based data cleaning method is used to remove duplicate, erroneous, and incomplete data records. At the same time, talent information is standardized, such as uniformly coding skill names and educational levels.

[0042] Use data deduplication technology based on hash algorithm to ensure the uniqueness of each record in the talent database;

[0043] The talent database constructed supports multi-dimensional retrieval and analysis of talent information.

[0044] S2. Then collect the company's recruitment demand text, historical recruitment data and industry trend information, and pre-process the demand data;

[0045] For this specific implementation method, the method of collecting corporate recruitment demand texts is: receiving recruitment demand texts submitted by enterprises through the corporate user terminal or the online platform of the talent digital brain system; corporate users can submit recruitment needs in various forms, such as filling out forms online, uploading Word documents or PDF files, etc., to improve user convenience.

[0046] Methods for collecting historical recruitment data: Extract historical recruitment data from the company's talent management system, including job information, candidate information, interview records, hiring results, etc.

[0047] Methods for collecting industry trend information: Use web crawler technology or third-party data service providers to collect industry trend information related to the recruitment position, such as industry development trends, emerging technology dynamics, market demand changes, etc.

[0048] The extracted data were denoised and normalized to remove outliers and irrelevant information to ensure the comparability and consistency of the data.

[0049] Based on specific needs, feature extraction and dimensionality reduction are performed on the required data to improve the efficiency and accuracy of subsequent analysis.

[0050] S3. Use natural language processing technology to analyze pre-processed demand data and extract core indicators such as key skills, experience requirements, and cultural fit;

[0051] For this specific implementation, a BERT or Transformer model is used to parse the pre-processed demand data to extract key skills, experience requirements, and cultural fit.

[0052] S4. Combining the company's historical data with industry dynamics, assign weights to each demand and use association rule mining algorithms to identify potential needs that the company has not clearly stated;

[0053] For this specific implementation method, combining the company's historical data and industry dynamics, the specific method of assigning weights to various requirements is as follows:

[0054] Combine the company's historical data with industry dynamics and dynamically adjust the weights of various requirements based on AHP or entropy weight method;

[0055] At the same time, the TF-IDF algorithm is used to evaluate the importance of industry trend keywords.

[0056] Combining the company's historical recruitment data, industry trend information and current recruitment demand text, we use association rule mining algorithms to analyze the correlation between various demands, thereby identifying potential demands that may not be clearly stated by the company but are more important in the actual recruitment process.

[0057] For example, if a company has repeatedly emphasized the importance of a certain skill or experience in historical recruitment, and this skill or experience is highly relevant to the responsibilities of the current recruitment position or industry trends, then the association rule mining algorithm can identify this potential demand and incorporate it into the company's recruitment demand set to improve the accuracy and effectiveness of recruitment.

[0058] S5. Integrate explicit demand and potential demand to form a structured set of corporate recruitment needs;

[0059] In this specific implementation, explicit requirements (such as key skills and experience requirements) extracted through natural language processing are first organized and categorized to form a preliminary requirements framework. Then, potential requirements (such as those not explicitly stated by the company but important in the actual recruitment process) identified through association rule mining algorithms are added to the requirements framework, and these potential requirements are clearly defined and described.

[0060] During the integration process, we combine historical company data with industry trends to weight each requirement to ensure the accuracy and effectiveness of the recruitment demand set. Weights can be dynamically adjusted using methods such as the Analytic Hierarchy Process (AHP) or the Entropy Weight Method. We also use the TF-IDF algorithm to assess the importance of industry trend keywords to reflect changes and trends in a company's recruitment needs.

[0061] Finally, the sorted explicit and potential needs are organized according to a certain logical structure and format to form a structured set of corporate recruitment needs. This set of needs will serve as the basis for subsequent multi-dimensional matching algorithms, used to intelligently screen candidates and generate personalized recommendation reports.

[0062] S6. Based on the talent digital brain system and combined with the company's recruitment needs set, a multi-dimensional matching algorithm is used to intelligently screen candidates. After the screening is completed, the system generates a personalized recommendation report.

[0063] In this specific implementation, based on the feature type of the enterprise recruitment demand set, cosine similarity, Euclidean distance or Manhattan distance is selected to calculate the similarity with the talents in the talent database to obtain a candidate set;

[0064] According to the weight of each feature in the recruitment demand set, the similarity of candidates in the candidate set is weighted summed;

[0065] The comprehensive matching degree of the candidates is obtained by weighted summation, and the candidates are ranked based on the comprehensive matching degree. The candidates with higher matching degrees are ranked higher.

[0066] After the system completes candidate screening and generates a personalized recommendation report:

[0067] Use the K-means clustering algorithm to group candidates; evaluate the matching degree through decision tree or random forest models; and generate a visual report containing talent profiles, skill tags, and matching scores.

[0068] Among them, feedback from companies and candidates is collected, categorized, and filed according to company, candidate, and feedback type;

[0069] Use data analysis tools and methods to deeply explore the reasons behind feedback data;

[0070] Based on the problem diagnosis results, the multi-dimensional matching algorithm is optimized in a targeted manner.

[0071] When collecting feedback from companies and candidates,

[0072] For enterprises: A dedicated feedback portal is set up in the Talent Digital Brain system, allowing enterprises to submit their evaluations of recommended candidates and adjust recruitment needs at any time;

[0073] For the talent side: After the candidates complete the interview, an evaluation link will be sent via SMS or email to collect their feedback on the interview process, the matching degree of the recommended position, and their own competitiveness assessment.

[0074] Collect candidates' comments and discussions on social media and recruitment platforms to understand their views and experiences with the positions recommended by the system.

[0075] As the talent database and corporate recruitment needs are continuously updated, the matching algorithm is rerun regularly to obtain the latest recommendation results.

[0076] The intelligent identification method based on the talent digital brain system of the present invention is used. When it is used specifically, a talent database is first constructed based on the talent digital brain system; then the company's recruitment demand text, historical recruitment data and industry trend information are collected, and the demand data is preprocessed; natural language processing technology is used to parse the preprocessed demand data, and core indicators such as key skills, experience requirements and cultural fit are extracted; based on the company's historical data and industry dynamics, weights are assigned to each demand, and the potential demand that is not clearly stated by the company is identified through association rule mining algorithms; explicit needs and potential needs are integrated to form a structured set of corporate recruitment needs; based on the talent digital brain system and combined with the company's recruitment demand set, a multi-dimensional matching algorithm is used to intelligently screen candidates. After the screening is completed, the system generates a personalized recommendation report. In this way, the technical problem that it is difficult to accurately match the actual needs of the company in the current talent recruitment and management work, resulting in low recruitment efficiency and waste of precious talent resources is solved.

[0077] Through natural language processing technology, the present invention can accurately extract core indicators such as key skills, experience requirements and cultural fit from the company's recruitment demand text, avoiding the errors and subjectivity of traditional manual analysis.

[0078] By using association rule mining algorithms, combined with corporate historical data and industry dynamics, we can identify potential needs that have not been clearly expressed by the company, further enrich the dimensions of recruitment needs, and make recruitment more in line with the actual needs of the company.

[0079] Based on a structured set of corporate recruitment needs, we use a multi-dimensional matching algorithm to intelligently screen candidates. This not only takes into account the candidate's skill matching, but also comprehensively considers multiple dimensions such as experience and cultural fit, greatly improving the accuracy and efficiency of recruitment.

[0080] At the same time, integrating explicit demand with potential demand to form a structured set of corporate recruitment needs makes recruitment needs clearer and more specific, helps companies evaluate candidates more accurately, and avoids waste of talent due to unclear needs.

[0081] The system generates personalized recommendation reports based on the screening results, providing companies with intuitive and easy-to-understand candidate evaluation results, helping them quickly locate the most suitable candidates and further optimize talent resource allocation.

[0082] The present invention greatly improves recruitment efficiency, shortens recruitment cycle and reduces recruitment costs by accurately matching actual needs of enterprises and optimizing talent resource allocation.

[0083] Personalized recommendation reports and precise recruitment needs help companies attract more outstanding talents and enhance their overall competitiveness and market position.

[0084] The above disclosure is only a preferred embodiment of the present invention, and certainly cannot be used to limit the scope of the rights of the present invention. Ordinary technicians in this field can understand that all or part of the processes of the above embodiment and equivalent changes made in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. An intelligent identification method based on a talent digital brain system, characterized in that: The steps include: First, build a talent database based on the talent digital brain system; Then collect the company's recruitment demand text, historical recruitment data and industry trend information, and pre-process the demand data; Use natural language processing technology to analyze pre-processed demand data and extract core indicators such as key skills, experience requirements, and cultural fit; Combining historical enterprise data with industry trends, we assign weights to various needs and use association rule mining algorithms to identify potential needs that are not clearly stated by the enterprise. Integrate explicit demand and potential demand to form a structured set of corporate recruitment needs; Based on the talent digital brain system and combined with the company's recruitment needs set, a multi-dimensional matching algorithm is used to intelligently screen candidates. After the screening is completed, the system generates a personalized recommendation report.

2. The intelligent identification method based on the talent digital brain system according to claim 1, characterized in that: The specific method of building a talent database based on the Talent Digital Brain System is as follows: The Talent Digital Brain System builds a talent database by integrating multi-source heterogeneous data; Specifically, a rule-based data cleaning method is used to remove duplicate, erroneous, and incomplete data records. At the same time, talent information is standardized, such as uniformly coding skill names and educational levels. Use data deduplication technology based on hash algorithm to ensure the uniqueness of each record in the talent database; The talent database constructed supports multi-dimensional retrieval and analysis of talent information.

3. The intelligent identification method based on the talent digital brain system according to claim 2, characterized in that: Use BERT or Transformer models to parse pre-processed demand data and extract key skills, experience requirements, and cultural fit.

4. The intelligent identification method based on the talent digital brain system according to claim 3, characterized in that: Combining the company's historical data with industry dynamics, the specific method for assigning weights to various requirements is as follows: Combining the company's historical data with industry dynamics, the weights of various requirements are dynamically adjusted based on AHP or entropy weight method; at the same time, the TF-IDF algorithm is used to evaluate the importance of industry trend keywords.

5. The intelligent identification method based on the talent digital brain system according to claim 4 is characterized in that: Based on the feature type of the company's recruitment demand set, select cosine similarity, Euclidean distance, or Manhattan distance to calculate the similarity with the talents in the talent database to obtain a candidate set; According to the weight of each feature in the recruitment demand set, the similarity of candidates in the candidate set is weighted summed; The comprehensive matching degree of the candidates is obtained by weighted summation, and the candidates are ranked based on the comprehensive matching degree. The candidates with higher matching degrees are ranked higher.

6. The intelligent identification method based on the talent digital brain system according to claim 5, characterized in that: After the system completes candidate screening and generates a personalized recommendation report: Use the K-means clustering algorithm to group candidates; evaluate the matching degree through decision tree or random forest models; and generate a visual report containing talent profiles, skill tags, and matching scores.

7. The intelligent identification method based on the talent digital brain system according to claim 6, characterized in that: Collect feedback from companies and candidates, classify and organize it, and file it according to company, candidate, and feedback type; Use data analysis tools and methods to deeply explore the reasons behind feedback data; Based on the problem diagnosis results, the multi-dimensional matching algorithm is optimized in a targeted manner.

8. The intelligent identification method based on the talent digital brain system according to claim 7, characterized in that: When collecting feedback from companies and candidates, For enterprises: A dedicated feedback portal is set up in the Talent Digital Brain system, allowing enterprises to submit their evaluations of recommended candidates and adjust recruitment needs at any time; For talent: After candidates complete their interviews, we send them an evaluation link via SMS or email to collect their feedback on the interview process, the recommended job fit, and their own competitiveness. Collect candidates' comments and discussions on social media and recruitment platforms to understand their views and experiences with the positions recommended by the system.

9. The intelligent identification method based on the talent digital brain system according to claim 8, characterized in that: As the talent database and corporate recruitment needs are continuously updated, the matching algorithm is rerun regularly to obtain the latest recommendation results.

Images