Rule set constraint-based enterprise human resource auxiliary decision-making method and system
Through the enterprise human resources-assisted decision-making method based on rules-set constraints, the problem that candidate resumes in traditional human resources management is solved, and a more efficient and objective candidate evaluation and decision-making is achieved.
Patent Information
- Application Number
- CN202411809101.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-10
- Publication Date
- 2025-04-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In traditional enterprise human resources management, human resources managers need to manually compare candidate resumes, which is time-consuming and labor-intensive and vulnerable to subjective factors, resulting in inconsistent decision-making.
A corporate human resources-assisted decision-making method based on rule set constraints is adopted. By collecting multi-dimensional personal information of candidates, filtering and scoring is performed based on the pre-rule set and the main rule set, comprehensive evaluation and final score are generated to reduce subjective bias.
It significantly reduces the time spent by HR managers on screening and comparing candidate resumes, improves decision-making efficiency, and ensures the objectivity and consistency of candidate competency assessments.
Smart Images
Figure CN119831550A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of data analysis, and in particular to an enterprise human resources decision-making assistance method and system based on rule set constraints. Background Art
[0002] In traditional corporate human resource management, human resource managers usually need to manually compare the resumes of multiple candidates in the management system to subjectively assess their suitability for a specific position. This process requires switching back and forth between the resumes of multiple candidates and comparing multiple fields of information one by one, which is time-consuming and labor-intensive. Moreover, when comparing the resumes of candidates, it is easy to be influenced by subjective factors and make one-sided evaluations of candidates; the weights of each dimension vary from person to person, resulting in inconsistent decisions. Summary of the invention
[0003] In order to make a more objective evaluation of candidates in combination with resume information, the present invention provides an enterprise human resources decision-making assistance method and system based on rule set constraints.
[0004] The present invention provides an enterprise human resources decision-making assistance method based on rule set constraints, which adopts the following technical solutions: An enterprise human resources decision-making assistance method based on rule set constraints includes the following steps: Collect personal information of candidates in multiple dimensions; Perform initial screening of personal information based on a pre-set of rules; Generate prompt words based on personal information and main rule set; The large model calculates dimension scores for personal information based on the prompt words and generates a comprehensive evaluation; Calculate the final score of personal information; Visual display of personal information scores and output of recommendations.
[0005] In a specific implementation scheme, personal information includes basic background, educational background, work background, project background, social relationships, and social evaluations.
[0006] In a specific feasible implementation method, after the personal information is initially screened according to the pre-rule set, the number of screening results is compared with the preset expected value. If the number of screening results is less than the expected value, a prompt to modify the pre-rule set is output, and the personal information is screened after obtaining the new pre-rule set until the number of screening results is greater than or equal to the expected value.
[0007] In a specific feasible implementation scheme, the main rule set includes a set of rules, secondary rules and weights. The primary rules include basic background, educational background, work background, project background, social relationships and social evaluation. The secondary rules are sub-dimensions under the dimensions of the primary rules.
[0008] In a specific implementation scheme, the calculation formula for calculating the final score of personal information is:
[0009] In the above formula, is the number of first-level rule fields; is the number of secondary rule fields; For the The weight of the first-level rule field; For the The first level rule The weight of the secondary rule field; For the The first level rule The score of the prompt word of the second-level rule.
[0010] In a specific implementation plan, the basic background includes the following dimensions: gender, ethnicity, birthplace, native place, age, marital status, health status, professional skills, specialties and hobbies; Educational background includes the following dimensions: academic level, graduation school, major, school location, academic performance, and community activities; Work background includes the following dimensions: years of work experience, work unit, position, title, work content, work location, and skill certificates; Project background includes the following dimensions: project location, project field, project type, roles undertaken, and project work duration; Social relations include the following dimensions: family members, social evaluation, organizational evaluation, personality test, and psychological test; Social evaluation includes the following dimensions: scientific research results, performance results, credit evaluation, rewards, and punishments.
[0011] In a specific possible implementation scheme, before the initial screening of personal information according to the pre-rule set, the previous personal information screening results are used as a sample set for model training generated by the pre-rule set and the main rule set.
[0012] In a specific possible implementation scheme, when visually displaying the personal information score, a preset marking color is used to mark the content associated with the prompt word in the personal information.
[0013] The present invention also provides an enterprise human resources decision-making support system based on rule set constraints, which adopts the following technical solution: An enterprise human resources decision-making support system based on rule set constraints, used to execute the above-mentioned enterprise human resources decision-making support method based on rule set constraints, comprises: Data collection unit, used to collect multi-dimensional personal information of candidates; The rule set generator is used to generate the pre-rule set and the main rule set according to actual needs; Large model prompt word generator, used to generate demand-oriented prompt words; Scoring engine, which accumulates scores for personal information based on prompt words; Visual display module, used to display the candidate's scores and marked personal information; Decision support system is used to comprehensively evaluate and output personal information.
[0014] In summary, the present invention has the following beneficial effects: Based on the multi-dimensional employee personal information data, the candidate data field weights were decomposed and the corresponding rule set was generated. The rule set can fully reflect user needs, conduct self-learning through data sources, and continuously iterate the rule set, ultimately achieving high adaptability of the rule set. It has the function of comparing multiple fields of candidate resumes at the same time, which significantly reduces the time spent by human resources management personnel on screening and comparing candidate resumes, and improves decision-making efficiency; the large model scores multiple candidates according to the adapted rule set, and accumulates the total score to generate a comprehensive evaluation of the candidates, provide decision-making suggestions, and ensure that the assessment of candidate capabilities is more objective and consistent, reducing the impact of subjective bias and enhancing the objectivity of decision-making. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 It is a flow chart of the enterprise human resources decision-making support method based on rule set constraints. DETAILED DESCRIPTION
[0016] The following combination Figure 1 The present invention is described in further detail.
[0017] Reference Figure 1 ,The enterprise human resource decision-making support method based on rule set constraints includes the following steps: S100, collecting personal information of candidates in multiple dimensions.
[0018] Personal information is extracted and collected from the human resources management system database and the Internet. Personal information includes gender, date of birth, ethnicity, native place, place of birth, time of employment, health status, marital status, professional expertise, language skills, credit record and other basic personal information and educational background, work experience, project experience, professional and technical titles, professional qualification certificates, rewards and punishments, social background, personality and psychological test results, and other people's evaluations.
[0019] Specifically, by connecting to the company's existing human resources management system database through an API, personal information on candidates in various dimensions, including basic background, educational background, work background, project background, social relationships, and social evaluations, is automatically extracted.
[0020] By connecting with other management information systems within the enterprise, relevant information about the candidate can be extracted. For example, the candidate's scientific research results information can be crawled from the enterprise's science and technology innovation management system. If the candidate is a party member, the candidate's organizational evaluation information can be extracted from the enterprise's party member management system.
[0021] Use web crawlers to crawl candidates' social evaluation information from their professional social platforms such as LinkedIn and GitHub, and obtain candidates' credit evaluation information from the Credit China website.
[0022] After personal information is collected, the personal information data will be checked for integrity and timeliness and stored.
[0023] S200, performing a preliminary screening of the candidates according to the pre-set rule set.
[0024] By writing various demand descriptions in the preset template, the requirements are generated to form basic requirements, primary rules and secondary rules. The basic requirements are written into the pre-set rule set, and the primary rules and secondary rules are written into the main rule set.
[0025] Personal information is screened according to the pre-set rule set, and personal information that meets the basic requirements of the pre-set rule set is retained. If the amount of retained personal information is less than the preset expected value, a prompt to modify the pre-set rule set is output, and the pre-set rule set is manually modified and screened again until the amount of personal information is greater than or equal to the expected value; otherwise, the next process is directly carried out.
[0026] S300, generating prompt words according to the personal information and the main rule set.
[0027] The first-level rules and second-level rules include initial weights and scoring rules. The first-level rules include dimensions such as basic background, educational background, work background, project background, social relationships, and social evaluation. The dimensions of the second-level rules are sub-dimensions under the first-level rule dimensions. Specifically: The basic background includes the following dimensions: gender, ethnicity, place of birth, native place, age, marital status, health status, professional skills, specialties and hobbies, etc.
[0028] Educational background includes the following dimensions: academic level, graduation school, major, school location, academic performance, club activities, etc.
[0029] Work background includes the following dimensions: years of work experience, work unit, position, professional title, work content, work location, skill certificates, etc.
[0030] Project background includes the following dimensions: project location, project field, project type, roles undertaken, project working time, etc.
[0031] Social relationships include the following dimensions: family members, social evaluations, organizational evaluations, personality tests, psychological tests, etc.
[0032] Social evaluation includes the following dimensions: scientific research results, performance results, credit evaluation, rewards, punishments, etc.
[0033] Table 1
[0034] Table 1 shows the first-level rule fields of the first-level rules and the second-level rule fields of the second-level rules, as well as the corresponding weight assignments. The data in Table 1 is only for reference, and the specific content and parameters will be adjusted according to actual needs. It is understandable that the previous personal information screening results can be used as a sample set for model training to make the generation of the pre-rule set and the main rule set more accurate, and the corresponding main rule set can be generated when facing the same needs in the future.
[0035] Furthermore, according to the primary rules and secondary rules in the main rule set, prompt words that can be recognized by the large model are extracted from the personal information, and the prompt words are associated with the primary rule fields and the secondary rule fields, which can reflect the matching degree of the personal information to the primary rules and the secondary rules.
[0036] At S400, the large model scores the dimensions of the personal information based on the prompt words and generates a comprehensive evaluation.
[0037] The big model analyzes personal information of different dimensions according to the prompt words, and then scores them by combining the first-level rules and the second-level rules to obtain the dimension scores. The big model further analyzes the personal information according to the dimension scores to generate a comprehensive evaluation of the candidate.
[0038] S500, calculating the final score of the personal information.
[0039] Specifically, for each prompt word, the weight of the corresponding secondary rule field is recorded as , the score output by the large model is recorded as , the final score of personal information is , which is the candidate’s final score. The calculation formula is as follows:
[0040] In the above formula, is the number of first-level rule fields; is the number of secondary rule fields; For the The weight of the first-level rule field; For the The first level rule The weight of the secondary rule field; For the The first level rule The score of the prompt word of the second-level rule.
[0041] The big model is used to analyze the scores of personal information and prompt words to generate a comprehensive evaluation of the personal information.
[0042] S600, visual display of personal information scores and output of recommendations.
[0043] According to the scores of different dimensions of personal information, a bar chart or radar chart is generated to show the scores of different personal information in different dimensions. The content associated with the prompt word in the personal information is marked with a preset marking color.
[0044] Based on the final score of personal information and comprehensive evaluation, candidate recommendations are output.
[0045] Example: To facilitate understanding, the following examples are given to further illustrate.
[0046] S100, collect personal information of 15 candidates in multiple dimensions.
[0047] S200, performing a preliminary screening of the candidates according to the pre-set rule set.
[0048] Get the requirement: On-site project manager for the construction of pumped storage power station in Guangdong. Because there was a need to recruit an on-site project manager in the past, the system stores a demand description template for the recruitment of on-site project managers. The built-in on-site project manager demand template is automatically matched according to the keyword "on-site project manager", and the professional, location, and project background information of the demand description are automatically filled and updated according to "Guangdong" and "pumped storage power station". The generated demand description example is shown in Table 2: Table 2
[0049] A pre-rule set as shown in Table 3 can be further generated.
[0050] Table 3
[0051] In the preliminary screening of the pre-set rule set, 3 out of 15 candidates passed the preliminary screening, which is greater than the expected value of 2. The personal information of the 3 candidates was subsequently processed.
[0052] S300, generating prompt words according to the personal information and the main rule set.
[0053] Generate the main rule set shown in Table 4.
[0054] Table 4
[0055] The project location is in Guangdong. Considering the language problem, the weight of R19's specialties and hobbies can be appropriately increased manually, and the weight of birthplace can be appropriately reduced accordingly. At the same time, the corresponding scoring rules are updated. The adjusted R19 scoring rules are shown in Table 5: Table 5
[0056] At S400, the large model scores the dimensions of the personal information based on the prompt words and generates a comprehensive evaluation.
[0057] The scoring results of the three candidates are shown in Table 6: Table 6
[0058] The comprehensive evaluation is shown in Table 7: Table 7
[0059] S500, calculating the final score of the personal information.
[0060] The final score for each candidate's personal information is as follows: Candidate A's final score: 87.5; Candidate B's final score: 78.3; Candidate C's final score: 72.1.
[0061] S600, visual display and suggestions of personal information scores.
[0062] The personal information of three candidates is displayed at the same time. The content related to the prompt words in the personal information is marked with different colors, and a chart is generated to show the candidates' scores in different dimensions.
[0063] Based on the final score and comprehensive evaluation, a decision recommendation is output: it is recommended to hire candidate A, who has the highest matching degree.
[0064] The present invention also discloses an enterprise human resources decision-making support system based on rule set constraints, which is used to execute the above-mentioned enterprise human resources decision-making support method based on rule set constraints, including: Data collection unit, used to collect multi-dimensional personal information of candidates; The rule set generator is used to generate the pre-rule set and the main rule set according to actual needs; Large model prompt word generator, used to generate demand-oriented prompt words; Scoring engine, which accumulates scores for personal information based on prompt words; Visual display module, used to display the candidate's scores and marked personal information; Decision support system is used to comprehensively evaluate and output personal information.
[0065] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A method for assisting enterprise human resources decision-making based on rule set constraints, characterized by: The steps include: Collect personal information of candidates in multiple dimensions; Perform initial screening of personal information based on a pre-set of rules; Generate prompt words based on personal information and main rule set; The large model calculates dimension scores for personal information based on the prompt words and generates a comprehensive evaluation; Calculate the final score of personal information; Visual display of personal information scores and output of recommendations.
2. The enterprise human resources decision-making support method based on rule set constraints according to claim 1 is characterized by: Personal information includes basic background, educational background, work background, project background, social relationships, and social evaluation.
3. The enterprise human resources decision-making support method based on rule set constraints according to claim 1 is characterized by: After the personal information is initially screened according to the pre-set rule set, the number of screening results is compared with the preset expected value. If the number of screening results is less than the expected value, a prompt to modify the pre-set rule set is output, and the personal information is screened after obtaining the new pre-set rule set until the number of screening results is greater than or equal to the expected value.
4. The enterprise human resources decision-making support method based on rule set constraints according to claim 1 is characterized by: The main rule set includes first-level rules, second-level rules and weights. The first-level rules include basic background, educational background, work background, project background, social relations and social evaluation. The second-level rules are sub-dimensions under the first-level rule dimensions.
5. The enterprise human resources decision-making support method based on rule set constraints according to claim 4 is characterized by: The calculation formula for calculating the final score of personal information is: In the above formula, is the number of first-level rule fields; is the number of secondary rule fields; For the The weight of the first-level rule field; For the The first level rule The weight of the secondary rule field; For the The first level rule The score of the prompt word of the second-level rule.
6. The enterprise human resources decision-making support method based on rule set constraints according to claim 4 is characterized by: Basic background includes the following dimensions: gender, ethnicity, birthplace, native place, age, marital status, health status, professional skills, specialties and hobbies; Educational background includes the following dimensions: academic level, graduation school, major, school location, academic performance, and community activities; Work background includes the following dimensions: years of work experience, work unit, position, title, work content, work location, and skill certificates; Project background includes the following dimensions: project location, project field, project type, roles undertaken, and project work duration; Social relations include the following dimensions: family members, social evaluation, organizational evaluation, personality test, and psychological test; Social evaluation includes the following dimensions: scientific research results, performance results, credit evaluation, rewards, and punishments.
7. The enterprise human resources decision-making support method based on rule set constraints according to claim 1 is characterized by: Before the initial screening of personal information according to the pre-rule set, the previous personal information screening results are used as a sample set for model training generated by the pre-rule set and the main rule set.
8. The enterprise human resources decision-making support method based on rule set constraints according to claim 1 is characterized by: When visually displaying the personal information score, a preset marking color is used to mark the content in the personal information that is associated with the prompt word.
9. An enterprise human resources decision-making support system based on rule set constraints, characterized by: The enterprise human resources decision-making assistance method based on rule set constraints for executing any one of claims 1 to 8 comprises: Data collection unit, used to collect multi-dimensional personal information of candidates; The rule set generator is used to generate the pre-rule set and the main rule set according to actual needs; Large model prompt word generator, used to generate demand-oriented prompt words; Scoring engine, which accumulates scores for personal information based on prompt words; Visual display module, used to display the candidate's scores and marked personal information; Decision support system is used to comprehensively evaluate and output personal information.