People and post matching method and device

Abstract

The embodiment of the invention provides a person and post matching method, and belongs to the technical field of data processing. The person and post matching method comprises the following steps: acquiring post description and multiple initial resumes of a target post; post information is obtained based on the post description, and resume information of each initial resume is obtained based on the multiple initial resumes; based on the resume information corresponding to the multiple initial resumes, the post information and a large language model, a target resume is determined from the multiple initial resumes, and the matching degree between the resume information of the target resume and the post information is larger than a preset threshold value. According to the technical scheme, the resumes and the posts are deeply analyzed through the large language model, more efficient and accurate person and post matching can be achieved, manual intervention is reduced, and recruitment efficiency is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of data processing, and in particular, to a method, device, computer device, computer-readable storage medium, and computer program product for person-job matching. Background Art

[0002] During the recruitment process, the human resources department (HR) faces a large amount of resume evaluation and screening work. However, the existing recruitment methods still have the following defects: (1) By manually reading and analyzing resumes to evaluate whether the job seekers of each resume are suitable for the recruitment position. This method not only has low efficiency but is also easily affected by subjective factors, resulting in the neglect of outstanding talents. (2) Through simple keyword matching, it is impossible to deeply analyze the resume and the job description, resulting in low recruitment accuracy.

[0003] It should be noted that the above content is not necessarily the prior art and is not used to limit the patent protection scope of the present application. Summary of the Invention

[0004] The embodiments of the present application provide a method, device, computer device, computer-readable storage medium, and computer program product for person-job matching to solve or alleviate one or more of the above technical problems.

[0005] One aspect of the embodiments of the present application provides a method for person-job matching, and the method includes: Obtain the job description of the target position and multiple initial resumes; Obtain the position information based on the job description, and obtain the resume information of each of the initial resumes based on the multiple initial resumes; Based on the resume information corresponding to each of the multiple initial resumes, the position information, and the large language model, determine the target resume from the multiple initial resumes, and the matching degree between the resume information of the target resume and the position information is greater than a preset threshold.

[0006] Optionally, obtaining the position information based on the job description, and obtaining the resume information of each of the initial resumes based on the multiple initial resumes includes: Parse the job description through natural language processing technology, identify and extract the position information from the job description, and the position information includes the target intervals corresponding to multiple features; Parse the initial resume through natural language processing technology, identify and extract the resume information from the initial resume, and the resume information includes the true values corresponding to the multiple features; Wherein, the natural language processing technology includes one or more of word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, and sentiment analysis.

[0007] Optionally, determining a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the position information, and a large language model includes: Matching the resume information corresponding to each of the multiple initial resumes with the position information; Determining the initial resumes whose true values of the multiple features are all within the corresponding target intervals as the pre-screened resumes; Inputting the resume information of the pre-screened resumes and the position information into the large language model to determine the target resume from the pre-screened resumes through the large language model.

[0008] Optionally, the large language model obtains the target resume through the following operations: Performing semantic recall based on the resume information of the pre-screened resumes and the position information to obtain the target resume; Wherein, the semantic recall includes: semantic analysis and semantic matching; and / or context analysis and context matching; the semantic matching degree and / or context matching degree between the resume information of the target resume and the position information is greater than the preset threshold.

[0009] Optionally, there are multiple target resumes, and the person-position matching method further includes: Obtaining a matching preference, where the matching preference includes one or more target features; Inputting the matching preference, the resume information corresponding to each of the multiple target resumes, and the position information into a pre-trained evaluation model to obtain a person-position matching list through the evaluation model; Pushing the person-position matching list to the target position and / or the target object corresponding to the target position; Wherein, the person-position matching list includes the multiple target resumes sorted by score, and the matching preference is used to guide the evaluation model: based on the matching degree between the true values of the one or more target features and the corresponding target intervals, score each of the target resumes.

[0010] Optionally, obtaining a matching preference includes: Based on the target industry corresponding to the target position, determining the industry preference degrees corresponding to the multiple features; Based on the industry preference degrees corresponding to the multiple features, determining the one or more target features from the multiple features.

[0011] Optionally, the person-job matching list further includes a resume analysis report corresponding to each of the multiple target resumes. The resume analysis report includes a summary of the target resume and the matching degree between the target resume and the target job. The resume analysis report is generated by the evaluation model based on the resume information of the target resume and the job information.

[0012] Another aspect of the embodiments of the present application provides a person-job matching device, which includes: A first acquisition module, configured to acquire a job description of a target job and multiple initial resumes; A second acquisition module, configured to acquire job information based on the job description, and acquire resume information of each of the multiple initial resumes based on the multiple initial resumes; A determination module, configured to determine a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the job information, and a large language model, where the matching degree between the resume information of the target resume and the job information is greater than a preset threshold.

[0013] Another aspect of the embodiments of the present application provides a computer device, including: At least one processor; and A memory communicatively connected to the at least one processor; Wherein: the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method as described above.

[0014] Another aspect of the embodiments of the present application provides a computer-readable storage medium, in which computer instructions are stored, and when the computer instructions are executed by a processor, the method as described above is implemented.

[0015] Another aspect of the embodiments of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, the method as described above is implemented.

[0016] The embodiments of the present application adopting the above technical solutions may include the following advantages: Acquire the job description of the target job, and acquire job information based on the job description. Acquire multiple initial resumes, and acquire resume information of each of the multiple initial resumes based on the multiple initial resumes. Determine a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the job information, and a large language model. Among them, the matching degree between the resume information of the target resume and the job information is greater than a preset threshold. It can be seen that the embodiments of the present application can achieve more efficient and accurate person-job matching, reduce manual intervention, and improve recruitment efficiency by deeply analyzing resumes and jobs through a large language model. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings exemplarily show embodiments and form part of the specification, and are used together with the written description of the specification to explain the exemplary embodiments of the embodiments. The shown embodiments are for illustrative purposes only and do not limit the scope of the claims. In all the drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0018] Figure 1 Schematically shows a flowchart of the person-job matching method according to Embodiment 1 of the present application; Figure 2 Schematically shows Figure 1 a sub-step flowchart of step S102 in; Figure 3 Schematically shows Figure 1 a sub-step flowchart of step S104 in; Figure 4 Schematically shows an additional flowchart of the person-job matching method according to Embodiment 1 of the present application; Figure 5 Schematically shows Figure 4 a sub-step flowchart of step S400 in; Figure 6 is an application example diagram of the person-job matching method according to Embodiment 1 of the present application; Figure 7 Schematically shows a block diagram of the person-job matching device according to Embodiment 2 of the present application; and Figure 8 Schematically shows a schematic diagram of the hardware architecture of a computer device according to Embodiment 3 of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the scope of protection of the present application.

[0020] It should be noted that in the embodiments of the present application, the descriptions involving "first", "second", etc. are only for descriptive purposes and should not be construed as indicating or implying their relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one such feature. Additionally, the technical solutions between various embodiments may be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement them. When the combination of technical solutions results in contradictions or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection required by the present application.

[0021] In the description of the present application, it should be understood that the numerical labels before the steps do not identify the order of execution of the steps, but are only used to facilitate the description of the present application and distinguish each step. Therefore, it should not be construed as a limitation to the present application.

[0022] First, the following provides the term explanations involved in the present application: NLP (Natural Language Processing): Natural Language Processing, which is a subfield of artificial intelligence, aims to enable computers to analyze, understand, interpret, and generate human language and interact with human language. NLP enables computers to process text data and extract valuable information from it.

[0023] NER (Named Entity Recognition): Named Entity Recognition, which is a technology in natural language processing, aims to identify named entities (such as person names, place names, organization names, etc.) in text and classify them. NER can be used in multiple fields such as information extraction, intelligence analysis, and search engines.

[0024] Machine Learning (ML): A branch of artificial intelligence, which refers to the process of enabling a computer to learn from data and make predictions or decisions through algorithms without explicit programming. Machine learning algorithms include decision trees, support vector machines (SVMs), neural networks, clustering, etc.

[0025] Deep Learning: A subfield of machine learning, which is based on a multi-layer architecture of neural networks (such as Convolutional Neural Network CNN, Recurrent Neural Network RNN, etc.) to automatically extract features and learn the representation of data. Deep learning can be used in multiple fields such as computer vision, natural language processing, and speech recognition.

[0026] Word Embedding: A technique in natural language processing used to map words into a high-dimensional vector space, enabling a computer to understand the semantic similarity between words. Word embedding methods include Word2Vec, GloVe, FastText, etc.

[0027] Model Training: The process in machine learning where a model learns from data and optimizes its parameters. The model is trained using a training set and adjusts its parameters by minimizing a loss function until it can make good predictions.

[0028] Inference: Refers to the process of making predictions using a trained model after the model training is completed.

[0029] Deep Learning Model: A machine learning model with multiple layers of neural networks that can automatically learn features from data and perform tasks such as classification, regression, and generation.

[0030] Recommendation System: A system that recommends suitable products, content, or services to a target object (such as a user) through data analysis.

[0031] CV (Curriculum Vitae): Resume.

[0032] JD (Job Description): Job description.

[0033] The embodiments of the present application provide a technical solution for person-job matching. In this technical solution: (1) Based on the large language model, the semantic information in the resume and job description is deeply analyzed, so as to accurately understand the comprehensive requirements of candidates (job seekers) and the position, thereby improving the accuracy of matching. Especially when dealing with recruitment in industries with high requirements or special requirements (such as the photovoltaic industry), it can ensure a high degree of fit between candidates and positions; (2) The matching solution can be customized and optimized based on the special needs of the industry (such as the photovoltaic industry), and combined with the professionalism of the industry (such as technical background, industry experience, preferences, etc.), to provide more accurate candidate recommendations; (3) Combining hard condition screening, semantic recall and precise matching, the most suitable candidates (the target objects corresponding to the target resumes) can be screened out in a short time, reducing the time and cost required for manual resume screening, and improving the automation and intelligence of the recruitment process. At the same time, through the scoring and ranking mechanism of the large language model, a list of the top N candidates can be provided for HR to help HR make quick decisions and further optimize the recruitment process; (4) For HR, there is no need to process a large number of resumes. The system can automatically generate a precise list of candidates and provide recommendations based on the comprehensive scores of the candidates. For job seekers, the system can accurately match positions that match their own backgrounds and skills, improving the job hunting experience. That is, reducing manual intervention and greatly improving the usage experience of HR and candidates, with obvious technical advantages and competitiveness; (5) Integrating hard condition screening and semantic matching to optimize the results of the preliminary screening and effectively narrow the range of candidates. See the following for details.

[0034] The technical solution of the present application will be introduced through multiple embodiments below. It should be noted that these embodiments can be implemented in various different forms and should not be construed as being limited only to the embodiments described herein.

[0035] Embodiment 1 Figure 1 A flowchart of the person-job matching method according to Embodiment 1 of the present application is schematically shown.

[0036] As Figure 1 shown, the person-job matching method may include steps S100 to S104, where: Step S100, obtaining the job description of the target position and multiple initial resumes.

[0037] Step S102, obtaining the position information based on the job description, and obtaining the resume information of each of the multiple initial resumes based on the multiple initial resumes.

[0038] Step S104, determining a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the position information, and the large language model, where the matching degree between the resume information of the target resume and the position information is greater than a preset threshold.

[0039] The person-job matching method provided in this embodiment obtains the job description of the target job and obtains job information based on the job description. Obtain multiple initial resumes and obtain the resume information of each initial resume based on the multiple initial resumes. Based on the resume information, job information, and large language model corresponding to each of the multiple initial resumes, determine the target resume from the multiple initial resumes. Among them, the matching degree between the resume information of the target resume and the job information is greater than the preset threshold. It can be seen that the embodiment of the present application can achieve more efficient and accurate person-job matching by deeply analyzing the resume and the job through the large language model, reduce manual intervention, and improve the recruitment efficiency.

[0040] The following combines Figure 1 , and elaborates on each step in steps S100 to S104 and other optional steps in detail.

[0041] Step S100, obtain the job description of the target job and multiple initial resumes.

[0042] The target job can be various jobs in various industries. The industries can be finance, healthcare, education, photovoltaic, etc. The person-job matching method provided in the embodiment of the present application is applicable to the recruitment in various industries and can be customized and optimized according to the special needs of the industry. Hereinafter, taking the photovoltaic industry as an example, an exemplary introduction to the person-job matching method will be given. The job description can be a detailed description of the target job, which can include job responsibilities, job requirements (such as required skills, work experience, education background, gender, age), work environment, etc. The job description can be written and published by the human resources department (HR) of the employer. The initial resume can be the material used by the job seeker to apply for the job, which can include a large amount of personal information, such as educational background, work experience, professional skills, project experience, etc. The initial resume can be a targeted resume submitted by the job seeker for the target job, or a general resume uploaded by the job seeker to the recruitment platform that is not targeted at a specific job.

[0043] Step S102, obtain job information based on the job description, and obtain the resume information of each initial resume based on the multiple initial resumes.

[0044] Job information can be the key information in the job description, which can include multi-dimensional job requirements (such as required skills, experience, education, etc.) and job conditions (such as salary, work location, working hours, work environment, etc.). Exemplarily, natural language processing techniques or large language models can be used to deeply analyze and understand the job description, and accurate job information can be extracted from the job description. Resume information can be the key information of job seekers, such as personal information, educational background, work experience, skills, etc. Similarly, natural language processing techniques or large language models can be used to deeply analyze and understand each initial resume to obtain the resume information of each initial resume. An exemplary solution is provided below.

[0045] In an alternative embodiment, as Figure 2 shown, step S102 may include: Step S200, parsing the job description through natural language processing techniques, identifying and extracting the job information from the job description, where the job information includes target intervals corresponding to respective features.

[0046] Step S202, parsing the initial resume through natural language processing techniques, identifying and extracting the resume information from the initial resume, where the resume information includes actual values corresponding to the respective features. Among them, the natural language processing techniques include one or more of word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, and sentiment analysis.

[0047] Exemplarily, natural language processing techniques can be used to perform word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, etc. on the text in the initial resume to accurately identify and extract resume information. Resume information can include actual values corresponding to respective features of job seekers. Among them, multiple features can correspond to multi-dimensional job requirements, and the features can be age, gender, work experience, etc. For example, resume information a can be extracted from initial resume A: 26 years old, male, two years of work experience, with photovoltaic design skills. Similarly, natural language processing techniques can be used to perform word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, etc. on the text in the job description to extract comprehensive and accurate job information. Job information can include target intervals corresponding to respective features, such as: 50 years old and below, gender not limited, with 3 years or more of work experience, bachelor's degree or above, and proficiency in photovoltaic design required. In some embodiments, the job information and resume information can also be structured for subsequent screening and matching.

[0048] In this embodiment, through natural language processing technology, the resume and the job position are deeply analyzed to extract key information, and the unstructured resume text and job description are converted into structured data that can be understood by machines, which can provide accurate data input for subsequent screening and matching. Through NLP technology, the core competitiveness of candidates and the core requirements of the target job position can be accurately identified, beyond simple keyword matching, to achieve a deeper level of understanding and analysis, and improve the accuracy of subsequent matching.

[0049] Step S104: Based on the resume information corresponding to each of the multiple initial resumes, the job position information, and the large language model, determine a target resume from the multiple initial resumes, where the matching degree between the resume information of the target resume and the job position information is greater than a preset threshold.

[0050] Exemplarily, the resume information corresponding to each of the multiple initial resumes and the job position information can be input into the large language model. The large language model can deeply analyze and semantically match the resume information and the job position information, and obtain a target resume from the multiple initial resumes whose matching degree with the job position information is greater than the preset threshold, so as to achieve accurate person-job matching. To improve the person-job matching effect, the person-job matching process can be further optimized. Multiple exemplary solutions are provided below.

[0051] In an alternative embodiment, as Figure 3 shown, step S104 may include: Step S300: Match the resume information corresponding to each of the multiple initial resumes with the job position information.

[0052] Step S302: Determine the initial resumes whose true values of the multiple features are all within the corresponding target intervals as the preliminarily screened resumes.

[0053] Step S304: Input the resume information of the preliminarily screened resumes and the job position information into the large language model to determine the target resume from the preliminarily screened resumes through the large language model.

[0054] Before inputting the resume information and the job information into the large language model, preliminary screening can be carried out first. Exemplarily, the hard conditions / criteria (i.e., the target ranges corresponding to multiple features) of the target job can be determined according to the job information, such as: bachelor's degree or above, two years of work experience, major in communication engineering, more than 3 years of work experience in the photovoltaic industry, etc. These hard conditions are unchangeable and must be met by candidates, and can be used to filter out candidates who do not meet the basic requirements. Specifically, the resume information of multiple initial resumes can be matched with the job information, and the initial resumes whose true values of multiple features are all within the corresponding target ranges are retained as the pre-screened resumes, and the initial resumes whose true value of any one feature does not meet the corresponding target range are excluded. In this way, semantic matching and hard condition screening can be integrated to obtain good preliminary screening results and effectively narrow down the candidate range. The hard condition screening can be implemented through a rule engine. A rule engine is a technology for automated decision-making that processes data through predefined rules. The rule engine can be used for simple condition matching and screening, and can quickly classify or make decisions on data. In some embodiments, the HR can also configure the hard conditions according to the actual situation (such as special industry requirements, etc.) and add them to the rule engine to further improve the quality of the preliminary screening and achieve industry customization. Input the resume information and the job information of the pre-screened resumes into the large language model, and the large language model can determine the target resume from the pre-screened resumes, improve the matching efficiency, and reduce the consumption of computing resources.

[0055] In this embodiment, semantic matching and hard condition screening are combined to exclude candidates who do not meet the basic requirements through simple rules, ensuring that only resumes that meet the job requirements are analyzed during subsequent matching, thus improving the matching efficiency.

[0056] In an alternative embodiment, the large language model obtains the target resume through the following steps: performing semantic recall based on the resume information of the pre-screened resume and the job information to obtain the target resume; wherein, the semantic recall includes: semantic analysis and semantic matching; and / or context analysis and context matching; the semantic matching degree and / or context matching degree between the resume information of the target resume and the job information is greater than the preset threshold.

[0057] Exemplarily, the large language model can perform semantic recall based on the resume information and job information of the preliminarily screened resumes to obtain target resumes. Semantic recall can include: semantic analysis and semantic matching; and / or context analysis and context matching. The target resume can be a preliminarily screened resume with a semantic matching degree and / or context matching degree between the resume information and the job information greater than a preset threshold. The large language model can perform semantic analysis on the resume information and job information to identify more implicitly matching information (such as skill requirements, industry experience, etc.) from them, achieving semantic-level matching. It is not limited to surface keyword matching and can deeply understand the similarity between the resume information and the job information. Through context analysis and context matching of the resume information and job information, candidates highly suitable for the target position can be more accurately identified, reducing common mis-matching problems in the recruitment system.

[0058] In this embodiment, through the deep understanding and careful comparison of the resume information and job information by the large language model, complex patterns and associations are identified, making up for the deficiencies of simple keyword matching, and more comprehensively evaluating whether candidates possess the implicit skills and experience required for the position, further optimizing the accuracy of person-job matching.

[0059] In an alternative embodiment, there are multiple target resumes. As Figure 4 shown, the person-job matching method can further include: Step S400, obtaining matching preferences, where the matching preferences include one or more target features.

[0060] Step S402, inputting the matching preferences, the resume information corresponding to each of the multiple target resumes, and the job information into a pre-trained evaluation model to obtain a person-job matching list through the evaluation model.

[0061] Step S404, pushing the person-job matching list to the target position and / or the target object corresponding to the target position. Among them, the person-job matching list includes the multiple target resumes sorted by score, and the matching preferences are used to guide the evaluation model: based on the matching degree between the true values of the one or more target features and the corresponding target intervals, score each of the target resumes.

[0062] Exemplarily, matching preferences can be obtained. The matching preferences may include one or more features with relatively high importance, such as: preference for majors in electronic information engineering, preference for master's degrees, etc. The matching preferences can represent the recruitment tendencies of the target positions. The matching preferences, the resume information corresponding to each of the multiple target resumes, and the position information are input into a pre-trained evaluation model. Among them, the evaluation model can be a large language model. The large language model can evaluate, score, and rank candidates based on multiple dimensions (i.e., multiple features). The scoring can comprehensively consider the potential of job seekers, the recruitment tendencies of the target positions (one or more target features), and the fit between job seekers and the target positions, to obtain a more accurate, intuitive, and reliable matching result, that is, a person-position matching list. The person-position matching list can include multiple target resumes sorted by score. In an alternative embodiment, the person-position matching list can further include a resume analysis report corresponding to each of the multiple target resumes. The resume analysis report can include a summary of the target resume and the matching degree between the target resume and the target position. The resume analysis report is generated by the evaluation model based on the resume information and the position information of the target resume. Further, the person-position matching list can be pushed to the target objects corresponding to the target resumes and / or the target positions, that is, job seekers and / or HRs. For job seekers, they can obtain positions that are precisely matched to their own backgrounds and skills. For HRs, they can quickly make decisions based on the scoring, ranking, and analysis results (resume analysis reports) of the large language model. Among them, the resume analysis report can provide a summary of the work experience, skills, educational background, etc. of job seekers, helping HRs understand the advantages and disadvantages of job seekers and further optimize the decision-making process. In some embodiments, the person-position matching list can be composed of target resumes with the top N scores, effectively simplifying the recommendation results and saving time.

[0063] In this embodiment, scoring and ranking through a large language model can comprehensively evaluate each dimension, improving the scientificity and accuracy of the matching results. Pushing the person-position matching list to the target objects can provide recruitment decision-making support for HRs. Through the detailed resume analysis report, HRs can quickly understand the advantages and suitability of each candidate, ensuring the rationality and efficiency of candidates.

[0064] In an alternative embodiment, as Figure 5 shown, step S400 may include: Step S500, based on the target industry corresponding to the target position, determine the industry preference degrees corresponding to each of the multiple features.

[0065] Step S502, based on the industry preference degrees corresponding to each of the multiple features, determine the one or more target features from the multiple features.

[0066] Exemplarily, based on the industry where the target position is located (such as the photovoltaic industry), the industry preference degrees corresponding to each of the multiple features can be determined. For example: in the photovoltaic industry, work experience is more highly regarded, so the industry preference degree corresponding to work experience is higher. In the photovoltaic industry, the age requirement is relatively loose, so the industry preference degree corresponding to age is lower. Based on the industry preference degrees of the multiple features, one or more target features can be determined from the multiple features.

[0067] In this embodiment, the matching algorithm can be customized and optimized according to the special needs of the industry to meet the special needs of each industry for high-skilled talents.

[0068] To make the present application easier to understand, the following provides an exemplary application in combination with Figure 6 an exemplary application is provided.

[0069] S1: Parse the resume (CV) and extract key information (resume information).

[0070] S2: Parse the job description (JD) and identify the job requirements (job information).

[0071] S3: Based on the job information, perform a hard condition screening to achieve a preliminary screening and eliminate the non-conforming resumes.

[0072] S4: Obtain the list of screened resumes (preliminary screening resumes).

[0073] S5: Through semantic recall and exact matching, achieve a deep semantic matching between the resume information and the job information, and obtain the person-job matching result (target resume).

[0074] S6: Score and rank the target resume through a large language model. Based on the comprehensive score, generate a ranking to obtain the person-job matching list.

[0075] S7: Push the person-job matching list to the HR to provide decision-making support for the HR.

[0076] In this exemplary application: through in-depth parsing of the CV and JD, combined with hard condition screening, semantic recall and exact matching, using a large language model for comprehensive scoring, ranking and recommendation, finally provide an efficient and accurate person-job matching result for the HR in each industry.

[0077] Embodiment 2 Figure 7 The block diagram of the person-job matching device according to Embodiment 2 of the present application is schematically shown. The device can be divided into one or more program modules. One or more program modules are stored in the storage medium and executed by one or more processors to complete the embodiments of the present application. The program modules referred to in the embodiments of the present application refer to a series of computer program instruction segments that can complete specific functions. The following description will specifically introduce the functions of each program module in this embodiment. AsFigure 7 As shown in Figure 7 , the device 1000 may include: a first acquisition module 1100, a second acquisition module 1200, and a determination module 1300, where: The first acquisition module 1100 is configured to acquire a job description of a target position and multiple initial resumes; The second acquisition module 1200 is configured to acquire position information based on the job description, and acquire resume information of each of the multiple initial resumes based on the multiple initial resumes; The determination module 1300 is configured to determine a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the position information, and a large language model, where the matching degree between the resume information of the target resume and the position information is greater than a preset threshold.

[0078] As an optional embodiment, acquiring position information based on the job description, and acquiring resume information of each of the multiple initial resumes based on the multiple initial resumes includes: Parsing the job description through natural language processing technology, identifying and extracting the position information from the job description, where the position information includes target intervals corresponding to multiple features; Parsing the initial resume through natural language processing technology, identifying and extracting the resume information from the initial resume, where the resume information includes true values corresponding to the multiple features; Wherein, the natural language processing technology includes one or more of word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, and sentiment analysis.

[0079] As an optional embodiment, determining a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the position information, and a large language model includes: Matching the resume information corresponding to each of the multiple initial resumes with the position information; Determining the initial resumes whose true values of the multiple features are all within the corresponding target intervals as the initially screened resumes; Inputting the resume information of the initially screened resumes and the position information into a large language model to determine the target resume from the initially screened resumes through the large language model.

[0080] As an optional embodiment, the large language model obtains the target resume through the following operations: Performing semantic recall based on the resume information of the initially screened resumes and the position information to obtain the target resume; Among them, the semantic recall includes: semantic analysis and semantic matching; and / or context analysis and context matching; the semantic matching degree and / or context matching degree between the resume information of the target resume and the position information is greater than the preset threshold.

[0081] As an optional embodiment, there are multiple target resumes, and the apparatus 1000 is further configured to: Obtain a matching preference, where the matching preference includes one or more target features; Input the matching preference, the resume information corresponding to each of the multiple target resumes, and the position information into a pre-trained evaluation model, so as to obtain a list of person-position matches through the evaluation model; Push the list of person-position matches to the target position and / or the target object corresponding to the target position; Among them, the list of person-position matches includes the multiple target resumes sorted by score, and the matching preference is used to guide the evaluation model: based on the matching degree between the true value of the one or more target features and the corresponding target interval, score each of the target resumes.

[0082] As an optional embodiment, obtaining a matching preference includes: Based on the target industry corresponding to the target position, determine the industry preference degrees corresponding to the multiple features; Based on the industry preference degrees corresponding to the multiple features, determine the one or more target features from the multiple features.

[0083] As an optional embodiment, the list of person-position matches further includes a resume analysis report corresponding to each of the multiple target resumes, and the resume analysis report includes an abstract of the target resume and the matching degree between the target resume and the target position, and the resume analysis report is generated by the evaluation model based on the resume information of the target resume and the position information.

[0084] Embodiment III Figure 8 FIG. schematically shows a hardware architecture diagram of a computer device 10000 suitable for implementing the person-position matching method according to Embodiment III of the present application. In some embodiments, the computer device 10000 may be a terminal device such as a smart phone, a wearable device, a tablet computer, a personal computer, a vehicle-mounted terminal, a game console, a virtual device, a workbench, a digital assistant, a set-top box, a robot, etc. In other embodiments, the computer device 10000 may be a rack server, a blade server, a tower server or a cabinet server (including an independent server or a server cluster composed of multiple servers), etc. As Figure 8As shown, the computer device 10000 includes, but is not limited to: a memory 10010, a processor 10020, and a network interface 10030 that can communicate with each other through a system bus. Among them: The memory 10010 includes at least one type of computer-readable storage medium. The readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (such as SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 10010 can be an internal storage module of the computer device 10000, such as the hard disk or memory of the computer device 10000. In other embodiments, the memory 10010 can also be an external storage device of the computer device 10000, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., equipped on the computer device 10000. Of course, the memory 10010 can also include both the internal storage module and the external storage device of the computer device 10000. In this embodiment, the memory 10010 is generally used to store the operating system and various application software installed on the computer device 10000, such as the program code of the human-post matching method. In addition, the memory 10010 can also be used to temporarily store various data that have been output or will be output.

[0085] In some embodiments, the processor 10020 can be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other chips. The processor 10020 is generally used to control the overall operation of the computer device 10000, such as performing control and processing related to data interaction or communication with the computer device 10000. In this embodiment, the processor 10020 is used to run the program code stored in the memory 10010 or process data.

[0086] The network interface 10030 may include a wireless network interface or a wired network interface, which is generally used to establish a communication link between the computer device 10000 and other computer devices. For example, the network interface 10030 is used to connect the computer device 10000 to an external terminal through a network, and establish a data transmission channel and a communication link between the computer device 10000 and the external terminal. The network may be a wireless or wired network such as an enterprise intranet (Intranet), the Internet, the Global System of Mobile communication (GSM for short), Wideband Code Division Multiple Access (WCDMA for short), 4G network, 5G network, Bluetooth, Wi-Fi, etc.

[0087] It should be noted that Figure 8 Only the computer device with components 10010 - 10030 is shown, but it should be understood that it is not required to implement all the shown components, and more or fewer components can be alternatively implemented.

[0088] In this embodiment, the person-job matching method stored in the memory 10010 can also be divided into one or more program modules and executed by one or more processors (such as the processor 10020) to complete the embodiments of this application.

[0089] Embodiment 4 The embodiments of this application also provide a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the person-job matching method in the embodiments are implemented.

[0090] In this embodiment, the computer-readable storage medium includes flash memory, hard disks, multimedia cards, card-type memories (such as SD or DX memories, etc.), random access memories (RAM), static random access memories (SRAM), read-only memories (ROM), electrically erasable programmable read-only memories (EEPROM), programmable read-only memories (PROM), magnetic memories, magnetic disks, optical disks, etc. In some embodiments, the computer-readable storage medium may be an internal storage unit of a computer device, such as the hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., equipped on the computer device. Of course, the computer-readable storage medium may also include both the internal storage unit and the external storage device of the computer device. In this embodiment, the computer-readable storage medium is generally used to store the operating system and various application software installed on the computer device, such as the program code of the human-job matching method in the embodiment. In addition, the computer-readable storage medium can also be used to temporarily store various data that have been output or will be output.

[0091] Embodiment 5 The embodiment of the present application also provides a computer program product, including a computer program, which implements the method in the above embodiment when executed by a processor.

[0092] Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the embodiments of the present application can be implemented by a general-purpose computer device. They can be concentrated on a single computer device or distributed on a network composed of multiple computer devices. Optionally, they can be implemented by program codes executable by the computer device, so that they can be stored in a storage device and executed by the computer device. And in some cases, the steps shown or described can be executed in a different order from here, or they can be separately made into individual integrated circuit modules, or multiple modules or steps among them can be made into a single integrated circuit module to implement. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0093] It should be noted that the above are only the preferred embodiments of the present application, and do not limit the patent protection scope of the present application. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A person-job matching method, characterized in that: The method comprises: Obtain job descriptions and multiple initial resumes for target positions; Acquire job information based on the job description, and acquire resume information of each of the initial resumes based on the multiple initial resumes; Based on the resume information corresponding to each of the multiple initial resumes, the position information and the large language model, a target resume is determined from the multiple initial resumes, and the matching degree between the resume information of the target resume and the position information is greater than a preset threshold.

2. The method according to claim 1, characterized in that Acquiring job information based on the job description, and acquiring resume information of each of the initial resumes based on the multiple initial resumes, including: Parsing the job description by natural language processing technology, identifying and extracting the job information from the job description, the job information including target intervals corresponding to each of a plurality of features; Parsing the initial resume by natural language processing technology, identifying and extracting the resume information from the initial resume, wherein the resume information includes true values ​​corresponding to each of the multiple features; Among them, the natural language processing technology includes one or more of word segmentation, part-of-speech tagging, word embedding, syntactic parsing, named entity recognition, and sentiment analysis.

3. The method according to claim 2, characterized in that Determining a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the position information, and the large language model, includes: Matching the resume information corresponding to each of the multiple initial resumes with the position information; Determine the initial resume whose true values ​​of the plurality of features are all within the corresponding target interval as the initial screening resume; The resume information and the position information of the initially screened resumes are input into a large language model, so as to determine the target resume from the initially screened resumes through the large language model.

4. The method according to claim 3, characterized in that The large language model obtains the target resume by the following operations: Perform semantic recall based on the resume information of the initial screening resume and the position information to obtain the target resume; Among them, the semantic recall includes: semantic analysis and semantic matching; and / or context analysis and context matching; the semantic matching degree and / or context matching degree between the resume information of the target resume and the job information is greater than the preset threshold.

5. The method according to claim 2, characterized in that: The target resume includes multiple copies, and the person-job matching method further includes: Acquire matching preferences, wherein the matching preferences include one or more target features; Inputting the matching preference, the resume information corresponding to each of the plurality of target resumes, and the job information into a pre-trained evaluation model, so as to obtain a person-job matching list through the evaluation model; Pushing the person-job matching list to the target job and / or the target object corresponding to the target job; The person-job matching list includes the multiple target resumes sorted by scores, and the matching preference is used to guide the evaluation model: scoring each target resume based on the matching degree between the true value of the one or more target features and the corresponding target interval.

6. The method according to claim 5, characterized in that Get matching preferences, including: Based on the target industry corresponding to the target position, determine the industry preference corresponding to each of the multiple characteristics; The one or more target features are determined from the multiple features based on the industry preferences corresponding to each of the multiple features.

7. The method according to claim 5, characterized in that The person-job matching list also includes a resume analysis report corresponding to each of the multiple target resumes, the resume analysis report including a summary of the target resume and a matching degree between the target resume and the target job, and the resume analysis report is generated by the evaluation model based on the resume information of the target resume and the job information.

8. A person-job matching device, characterized in that: The device comprises: The first acquisition module is used to obtain the job description of the target position and multiple initial resumes; A second acquisition module, configured to acquire job information based on the job description, and to acquire resume information of each of the initial resumes based on the multiple initial resumes; A determination module is used to determine a target resume from the multiple initial resumes based on the resume information corresponding to each of the multiple initial resumes, the job information and the large language model, wherein the matching degree between the resume information of the target resume and the job information is greater than a preset threshold.

9. A computer device, characterized in that: include: at least one processor; and a memory communicatively connected to the at least one processor; wherein: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, the method according to any one of claims 1 to 7 is implemented.

11. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images