Intelligent recruitment and employment recommendation system based on big data

Abstract

The invention discloses an intelligent recruitment and employment recommendation system based on big data, belongs to the technical field of computers, and provides a multi-level and multi-dimensional intelligent matching system according to the technical scheme. According to the system, through the screening process of three stages, namely industry matching, enterprise matching and position matching, the candidate range is gradually reduced, and finally the most suitable job seeker is recommended. And a matching degree threshold is set in each stage, so that high quality of a recommendation result is ensured. Besides, the system makes full use of big data, not only considers skills and experiences of job seekers, but also incorporates multiple factors such as industry characteristics, enterprise culture and specific position requirements, and realizes more comprehensive and more accurate matching.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to an intelligent recommendation system for recruitment and employment based on big data. Background Art

[0002] Most traditional recruitment and employment platforms rely on job seekers to input resume information and companies to post job information. When matching, these platforms are often based on simple keyword matching, such as matching the job title with the work experience title in the resume. Some platforms have some basic screening functions, such as screening for education, years of work experience, etc., but these screening dimensions are relatively limited and cannot deeply explore the potential match between job seekers and positions.

[0003] The simple matching algorithms of existing recruitment platforms result in a large number of mismatches between job seekers and positions. For example, a job seeker with rich project management experience but a mismatched job title may be missed, and the company may miss out on suitable talent.

[0004] Therefore, how to recommend highly matched talents to enterprises has become a hot topic of research. Summary of the invention

[0005] The embodiment of the present application provides an intelligent recommendation system for recruitment and employment based on big data, which can improve the matching degree between recommended talents and enterprises, and at the same time, can improve efficiency compared with manual screening in the prior art. The technical solution is as follows:

[0006] Provided is a big data-based intelligent recommendation system for recruitment and employment, the system comprising:

[0007] an acquisition module, configured to, in response to an object recommendation request of a target enterprise, acquire object call information of the target enterprise, object information of a plurality of initial objects, industry information of the industry in which the target enterprise is located, and enterprise evaluation information of the target enterprise, wherein the plurality of initial objects are objects recommended by a plurality of target object recommendation platforms, wherein the plurality of target object recommendation platforms are object recommendation platforms whose credibility is greater than or equal to a credibility threshold, and the object call information includes enterprise information of the target enterprise and position information of a target position, wherein the target position is a position to be recommended;

[0008] A reference object determination module, configured to determine a plurality of reference objects from the plurality of initial objects based on the object information of the plurality of initial objects and the industry information, wherein the reference objects are initial objects whose matching degree with the industry of the target enterprise is greater than or equal to a first matching degree threshold;

[0009] a candidate object determination module, configured to determine a plurality of candidate objects from the plurality of reference objects based on the enterprise information of the target enterprise, the enterprise evaluation information of the target enterprise, and the object information of the plurality of reference objects, wherein the candidate objects are reference objects whose matching degree with the target enterprise is greater than or equal to a second matching degree threshold;

[0010] a target object determination module, configured to determine at least one target object from the plurality of candidate objects based on the position information of the target position and the object information of the plurality of candidate objects, wherein the target object is a reference object whose matching degree with the target position is greater than or equal to a third matching degree threshold;

[0011] A recommendation module is used to recommend the at least one target object to the target enterprise.

[0012] The technical solution provided in the embodiment of the present application proposes a multi-level and multi-dimensional intelligent matching system. The system gradually narrows the candidate range through a three-stage screening process, namely industry matching, enterprise matching and position matching, and finally recommends the most suitable job seeker. A matching threshold is set at each stage to ensure the high quality of the recommendation results. In addition, the system makes full use of big data, not only considering the skills and experience of job seekers, but also incorporating multiple factors such as industry characteristics, corporate culture and specific job requirements, to achieve a more comprehensive and accurate matching, and at the same time, compared with manual screening in the prior art, it can improve efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 It is a schematic diagram of an implementation environment of a big data-based intelligent recommendation system for recruitment and employment provided in an embodiment of the present application;

[0015] Figure 2 It is a structural diagram of an intelligent recommendation system for recruitment and employment based on big data provided in an embodiment of the present application;

[0016] Figure 3 is a structural diagram of a reference object determination module provided in an embodiment of the present application;

[0017] Figure 4 is a schematic diagram of the structure of a candidate object determination module provided in an embodiment of the present application;

[0018] Figure 5 It is a structural diagram of a target object determination module provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0020] In this application, the terms "first", "second", etc. are used to distinguish identical or similar items with basically the same effects and functions. It should be understood that there is no logical or temporal dependency between "first", "second", and "nth", nor is there any limitation on quantity and execution order.

[0021] Big data: Big data is a collection of data that is huge in scale, complex and diverse in types, and far exceeds the capabilities of traditional database software tools in terms of acquisition, storage, management, and analysis.

[0022] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain better results.

[0023] Machine Learning (ML) is a multi-disciplinary subject that involves probability theory, statistics, approximation theory, convex analysis, algorithm complexity theory and other disciplines. It specializes in studying how computers simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge sub-models to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to make computers intelligent. Its applications are spread across all areas of artificial intelligence. Machine learning and deep learning usually include artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and teaching learning.

[0024] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0025] Figure 1 This is a schematic diagram of the implementation environment of an intelligent recommendation method for recruitment and employment based on big data provided by the embodiment of the present application, see Figure 1 The implementation environment may include a first node 110, a second node 120 and a server 140.

[0026] The first node 110 is connected to the server 140 via a wireless network or a wired network. Optionally, the first node 110 is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The first node 110 is installed and runs an application that supports intelligent recommendation of recruitment and employment based on big data. In the embodiment of the present application, the first node 110 is a node corresponding to the target object recommendation platform, and the server 140 can obtain the object information of the initial object from the first node 110.

[0027] The second node 120 is connected to the server 140 via a wireless network or a wired network. Optionally, the second node 120 is a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The second node 120 is installed and runs an application that supports intelligent recommendation of recruitment and employment based on big data. In the embodiment of the present application, the second node 120 is a node corresponding to the target enterprise, and the server 140 can obtain the object call information of the target enterprise from the second node 120.

[0028] The server 140 is an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, distribution networks (Content Delivery Network, CDN) and big data and artificial intelligence platforms. The server 140 is deployed with the big data-based recruitment and employment intelligent recommendation system provided in the embodiment of the present application. The server 140 can provide background services for the applications running on the first node 110 and the second node 120. In the embodiment of the present application, the server is also referred to as an online modeling platform.

[0029] In traditional recruitment and employment platforms, the matching algorithm is too simple, mainly relying on keyword matching and basic screening functions. This method cannot deeply explore the potential matching between job seekers and positions, resulting in a large number of mismatches between job seekers and positions. Specifically, this simple matching method cannot take into account multi-dimensional factors such as the comprehensive ability of job seekers, career development potential, and compatibility with corporate culture. Therefore, how to achieve accurate matching between companies and job seekers based on big data has become a technical problem that needs to be solved urgently.

[0030] On a large Internet recruitment platform, hundreds of thousands of resumes and tens of thousands of job postings are posted every day. The platform uses a keyword-based matching algorithm, but its accuracy rate is only 30%. For example, a software engineer with rich project management experience was ignored by the system when matching the project manager position because the keyword "project manager" did not appear in his resume. At the same time, some cross-industry job seekers are difficult to be identified as potential matches by the system because their past work experience is different from the industry of the target position. In addition, the system cannot effectively consider soft factors such as corporate culture and work environment, resulting in even job seekers with matching skills may leave due to cultural incompatibility after joining the company.

[0031] If this technical problem cannot be effectively solved, it will cause serious waste of talent resources and inefficient corporate recruitment. Specifically, a large number of potentially suitable talents may be ignored by the system, while unsuitable candidates may be recommended to companies, increasing the screening costs of companies. For job seekers, they may miss many suitable job opportunities and extend the job search cycle. From a technical point of view, this problem will also lead to a decline in the user experience of the system and a decrease in user trust, which may ultimately affect the operation of the entire platform. Therefore, it is particularly important to develop an intelligent recommendation system that can analyze the matching degree of job seekers and positions in multiple dimensions and in depth.

[0032] In traditional recruitment and employment platforms, the matching algorithm is too simple, mainly relying on keyword matching and basic screening functions, and is unable to deeply explore the potential matching between job seekers and positions. This has led to a large number of job seekers and positions not matching, affecting recruitment efficiency and job search success rate.

[0033] To solve this problem, this application explored a variety of possible solutions. First, we considered adding more screening dimensions, such as education, years of work experience, skills, etc. However, although this method can improve the matching accuracy, it still cannot fully evaluate the matching degree between job seekers and positions.

[0034] Next, consider introducing artificial intelligence technologies, such as natural language processing and machine learning algorithms, to perform semantic analysis on resumes and job descriptions. This approach can better understand the abilities and experience of job seekers, but still lacks consideration of industry and corporate characteristics.

[0035] After in-depth thinking, this application realizes that a multi-level and multi-dimensional matching system is needed. This system should not only consider the personal information of job seekers, but also the industry characteristics, corporate culture and specific job requirements. Therefore, a recruitment and employment intelligent recommendation system based on big data is proposed.

[0036] Specifically, the system first obtains the target enterprise's object call information, the object information of multiple initial objects, the industry information of the target enterprise's industry, and the enterprise evaluation information of the target enterprise. Among them, the initial objects come from a highly credible recommendation platform to ensure the reliability of the data.

[0037] The system then gradually narrows down the candidates through a multi-step screening process:

[0038] First, based on the information of the initial candidates and the industry information, we screen out reference candidates who have a high degree of match with the target company's industry. This step ensures that the recommended job seekers match the company in terms of industry background.

[0039] Secondly, based on the information and evaluation of the target company, candidates with a high degree of match with the company are screened from the reference objects. This step takes into account the characteristics of the company and ensures that the recommended job seekers can adapt to the corporate culture and working environment.

[0040] Finally, based on the specific requirements of the target position, the final target candidates are determined from the candidates. This step ensures that the recommended job seekers are not only matched at the industry and company level, but also meet the requirements of the specific position.

[0041] Through this multi-level screening process, the system can comprehensively consider the matching degree between job seekers, enterprises and positions, greatly improving the accuracy and efficiency of matching. At the same time, by setting multiple matching thresholds, the system can retain the most suitable candidates in each step of screening, eliminate unsuitable candidates, and further improve the quality of recommendations.

[0042] The following is an introduction to the big data-based intelligent recommendation system for recruitment and employment (hereinafter referred to as the system) provided in the embodiments of the present application through several embodiments.

[0043] Embodiment 1:

[0044] Figure 2 This is a schematic diagram of the structure of a big data-based intelligent recommendation system for recruitment and employment provided in an embodiment of the present application. Figure 2 , the system 200 includes:

[0045] The acquisition module 201 is used to respond to the object recommendation request of the target enterprise, obtain the object calling information of the target enterprise, the object information of multiple initial objects, the industry information of the industry in which the target enterprise is located, and the enterprise evaluation information of the target enterprise, wherein the multiple initial objects are objects recommended by multiple target object recommendation platforms, and the multiple target object recommendation platforms are object recommendation platforms with a credibility greater than or equal to a credibility threshold. The object calling information includes the enterprise information of the target enterprise and the position information of the target position, and the target position is the position to be recommended.

[0046] Among them, the target recruitment information refers to the information released by the target enterprise for recruiting talents, which can be realized by enterprise information and position information. The initial object refers to the potential job seekers recommended by the recommendation platform with high credibility, which can be realized by the job seekers' personal information, work experience and educational background. Industry information refers to the relevant data of the industry in which the target enterprise is located, which can be realized by industry development trends, skill requirements and market prospects. Enterprise evaluation information refers to various evaluation data about the target enterprise, which can be realized by employee satisfaction, company culture and working environment.

[0047] The reference object determination module 202 is used to determine multiple reference objects from the multiple initial objects based on the object information of the multiple initial objects and the industry information, and the reference objects are initial objects whose matching degree with the industry of the target enterprise is greater than or equal to a first matching degree threshold.

[0048] Among them, the degree of matching refers to the fit between the job seeker and the company or position, which can be achieved by using a multi-dimensional scoring system.

[0049] The candidate object determination module 203 is used to determine multiple candidate objects from the multiple reference objects based on the enterprise information of the target enterprise, the enterprise evaluation information of the target enterprise and the object information of the multiple reference objects. The candidate objects are reference objects whose matching degree with the target enterprise is greater than or equal to a second matching degree threshold.

[0050] The target object determination module 204 is used to determine at least one target object from the multiple candidate objects based on the position information of the target position and the object information of the multiple candidate objects, and the target object is a reference object whose matching degree with the target position is greater than or equal to a third matching degree threshold.

[0051] The recommendation module 205 is used to recommend the at least one target object to the target enterprise.

[0052] The core innovation of this application is to propose a multi-level and multi-dimensional intelligent matching system. The system gradually narrows down the candidate range through a three-stage screening process, namely industry matching, enterprise matching and position matching, and finally recommends the most suitable job seekers. Matching thresholds are set at each stage to ensure the high quality of the recommendation results. In addition, the system makes full use of big data, not only considering the skills and experience of job seekers, but also incorporating multiple factors such as industry characteristics, corporate culture and specific job requirements, to achieve a more comprehensive and accurate match.

[0053] The working principle of this application can be described in detail as follows:

[0054] First, the system obtains various types of information. The target recruitment information is obtained from the target enterprise, including basic enterprise information and specific job requirements. The initial target information is obtained from a highly credible recommendation platform, including the job seeker's personal information, work experience, and educational background. Industry information is obtained from relevant databases or industry reports, including industry development trends and skill requirements. Enterprise evaluation information is obtained from channels such as employee feedback and public evaluations, including company culture and work environment evaluations.

[0055] The system then conducts a three-stage screening:

[0056] The first stage is industry matching. The system uses the initial object information and industry information to calculate the matching degree between each initial object and the target enterprise's industry. The matching degree calculation may involve factors such as the job seeker's industry experience and relevant skills. Only initial objects whose matching degree exceeds the first matching degree threshold will be selected as reference objects.

[0057] The second stage is enterprise matching. The system uses the target enterprise information, enterprise evaluation information and reference object information to calculate the matching degree between each reference object and the target enterprise. This stage may consider factors such as corporate culture and working environment. Only reference objects whose matching degree exceeds the second matching degree threshold will be selected as candidates.

[0058] The third stage is job matching. The system uses the target job information and candidate information to calculate the degree of match between each candidate and the target job. This stage may focus on specific skill requirements, work experience and other factors. Only candidates whose match exceeds the third match threshold will be finally selected as the target.

[0059] Finally, the system recommends the selected target objects to the target enterprise. The first matching degree threshold, the second matching degree threshold and the third matching degree threshold are set by the technician according to the actual situation, and the embodiment of the present application does not limit this.

[0060] This multi-stage screening design enables the system to comprehensively consider the matching degree between job seekers, companies and positions, greatly improving the accuracy and efficiency of matching. By setting multiple matching thresholds, the system can retain the most suitable candidates in each screening step and eliminate unsuitable candidates, further improving the quality of recommendations.

[0061] Embodiment 2:

[0062] The above-mentioned embodiment 1 is a brief introduction to the system provided by the embodiment of the present application. The system will be described below through other embodiments.

[0063] The reference object determination module 202 of the system is introduced. The object information includes historical work information, education information and object evaluation information. The industry information includes industry description information, industry development trend information and industry evaluation information. Figure 3 , the reference object determination module 202 includes:

[0064] The first industry matching degree determining unit 2021 is used to determine the first industry matching degree between each initial object and the industry based on the historical work information, education information and industry description information of each initial object.

[0065] The second industry matching degree determining unit 2022 is used to determine the second industry matching degree between each initial object and the industry based on the object evaluation information of each initial object and the industry development trend information.

[0066] The third industry matching degree determination unit 2023 is used to determine the third industry matching degree between each initial object and the industry based on the education information of each initial object, the object evaluation information and the industry evaluation information.

[0067] The first fusion unit 2024 is used to fuse the first industry matching degree, the second industry matching degree and the third industry matching degree between each initial object and the industry to obtain the matching degree between each initial object and the industry.

[0068] The first matching unit 2025 is used to determine the initial objects among the multiple initial objects whose matching degree with the industry is greater than or equal to the first matching degree threshold as reference objects, so as to obtain the multiple reference objects.

[0069] In a possible implementation, the technical solution of determining multiple reference objects from multiple initial objects based on the object information and industry information of multiple initial objects in the above embodiment 2. The object information includes historical work information, education information and object evaluation information, and the industry information includes industry description information, industry development trend information and industry evaluation information.

[0070] In a possible implementation, the reference object is determined by multi-dimensional matching calculation in Example 2. First, the first matching degree is calculated using the object's historical work information, education information and industry description information, reflecting the matching of the object's work experience and education background with the industry. Secondly, the second matching degree is calculated using the object evaluation information and industry development trend information, reflecting the degree of fit between the object's evaluation and the industry development direction. Thirdly, the third matching degree is calculated using education information, object evaluation information and industry evaluation information, reflecting the consistency of the object's education background and evaluation with the overall evaluation of the industry. Finally, these three matching degrees are combined to obtain the overall matching degree, and the final reference object is screened out based on the threshold.

[0071] As a specific example, suppose there is an initial object A, whose historical work information shows that he has worked as a software engineer in multiple Internet companies, and whose education information shows that he has a bachelor's degree in computer science. The object evaluation information mentions that he is "good at learning new technologies" and "has good teamwork skills." The target industry is the artificial intelligence industry, and the industry description information mentions that "solid programming foundation and machine learning knowledge are required." The industry development trend information points out that "deep learning and natural language processing are the focus of future development." The industry evaluation information emphasizes that "innovation and rapid learning are important."

[0072] When calculating the first industry match, the system will compare the correlation between A's software engineer experience and computer science degree and the description of the artificial intelligence industry, and may get a higher match, such as 0.8 (assuming the full score is 1). When calculating the second industry match, the system will analyze the fit between A's "good at learning new technologies" and the industry development trend, and may get a medium match, such as 0.6. When calculating the third industry match, the system will compare A's learning ability with the consistency of the rapid learning ability emphasized in the industry evaluation, and may get a higher match, such as 0.7.

[0073] Finally, the system may use a weighted average method to fuse these three matching degrees. Assuming the weights are 0.4, 0.3, and 0.3 respectively, the final matching degree is 0.80.4+0.60.3+0.7*0.3=0.71. If the first matching degree threshold is set to 0.7, the initial object A will be determined as the reference object.

[0074] Example 2 can effectively screen out reference objects with a high degree of match with the target industry from multiple initial objects. Compared with traditional methods that rely only on a single dimension (such as work experience), the method of the present application can more comprehensively evaluate the suitability of candidates. For example, even if a candidate has no direct experience in artificial intelligence work, if he has strong learning ability and relevant technical background, he may be identified as a potential suitable candidate. This method not only improves the accuracy of talent matching, but also expands the company's talent pool, which helps to discover potential outstanding talents.

[0075] In addition, the method provided in Example 2 is also highly flexible and scalable. By adjusting the weights of different dimensions or adding new matching dimensions, it can easily adapt to the special needs of different industries or enterprises. For example, for rapidly developing emerging industries, the weights of innovation ability and adaptability can be increased; for industries that require a lot of teamwork, the evaluation of communication skills and team spirit can be increased.

[0076] Embodiment 3:

[0077] The first industry matching degree determination unit 2021 is further introduced.

[0078] In a possible implementation, the historical work information includes historical work content, historical work enterprise and historical work industry, and the education information includes historical course information, historical school information, academic degree information and skill training information.

[0079] The first industry matching degree determination unit 2021 is used to determine the historical work matching degree between the initial object and the industry for any initial object among the multiple initial objects based on the historical work content, historical work enterprise, historical work industry and the industry description information of the initial object. Based on the academic degree information, skill training information and the industry description information of the initial object, the degree of academic qualifications and skills matching between the initial object and the industry is determined. The historical course information and historical school information of the initial object are used for query to obtain the historical course content and historical employment direction corresponding to the initial object. Based on the historical course content, historical employment direction and the industry description information corresponding to the initial object, the degree of direction matching between the initial object and the industry is determined. Based on the historical work matching degree, academic qualifications and skills matching degree and historical employment direction between the initial object and the industry, the first industry matching degree between the initial object and the industry is determined.

[0080] The above technical solution includes the following main processes:

[0081] In a possible implementation, the processing of historical work information: the historical work information includes historical work content, historical work enterprise and historical work industry. This information is used to determine the degree of historical work matching between the initial object and the industry. The historical work information can be compared with the industry description information through methods such as text analysis and semantic matching to obtain the degree of matching.

[0082] In a possible implementation, the processing of education information: education information includes history course information, history school information, degree information and skill training information. This information is used to determine the degree of matching between education and skills and the degree of matching between directions. By establishing an industry knowledge graph, the education information can be mapped with the knowledge and skills required by the industry to obtain the degree of matching.

[0083] In a possible implementation, analysis of history course content and history employment direction: by querying the history course information and history school information of the initial object, more detailed history course content and history employment direction can be obtained. This information can be analyzed by natural language processing technology, keywords and topics can be extracted, and compared with industry description information to obtain the degree of direction matching.

[0084] In a possible implementation, the fusion of multi-dimensional matching degrees: In the technical solution provided in Example 3, the historical work matching degree, the education and skills matching degree, and the direction matching degree are fused to obtain the final first industry matching degree. The fusion method can adopt weighted average, machine learning model, etc. to balance the importance of each dimension.

[0085] It should be noted that there are close connections and interactions between these features. For example, historical work information and education information may overlap, and appropriate weight distribution is needed to avoid repeated calculation of information. At the same time, the analysis results of historical course content and historical employment direction may have an impact on the degree of historical job matching and the degree of academic qualification and skill matching, and this interaction needs to be considered in the fusion process.

[0086] In Example 3, by comprehensively considering the historical work experience, educational background and industry information of the initial object, a comprehensive assessment of the matching degree between the initial object and the specific industry is achieved. Specifically, in Example 3, the historical work matching degree, the educational background and skill matching degree and the direction matching degree are calculated respectively, and then these matching degrees are combined to obtain the final first industry matching degree.

[0087] Compared with traditional simple keyword matching or single-dimensional matching, this multi-dimensional matching calculation method can more comprehensively and accurately evaluate the matching degree between the initial object and the industry. By considering the historical work content, historical work company and historical work industry, we can fully understand whether the initial object's work experience is relevant to the target industry. At the same time, by analyzing the educational background, including academic qualifications, skills training, historical course content and historical employment direction, we can evaluate whether the initial object's knowledge reserves and career development direction match the target industry.

[0088] The technical solution in Example 3 can be implemented as follows in practical applications:

[0089] First, for the historical work information of the initial object, natural language processing technology can be used to extract keywords and topics. For example, for a software developer, keywords such as "Java development", "database design", and "agile development" may be extracted. Then, these keywords are semantically matched with the industry description information and the matching degree is calculated.

[0090] For education information, an industry-discipline mapping table can be established to match academic degree information and skill training information with industry needs. For example, for the IT industry, computer science, software engineering and other related majors may be given a higher match.

[0091] When analyzing history course content and history career directions, text clustering algorithms can be used to classify course content and career directions into different topics, and then match them with industry description information. For example, if a person's course content contains a lot of data analysis-related courses, and the target industry is the big data industry, then a high degree of direction matching will be obtained.

[0092] Finally, when integrating the matching degrees of each dimension, a weighted average method can be used. For example, a weight of 0.4 can be given to the matching degree of historical work, a weight of 0.3 to the matching degree of education and skills, and a weight of 0.3 to the matching degree of direction. These weights can be adjusted according to actual conditions.

[0093] In this way, the technical solution in Example 3 can more comprehensively and accurately evaluate the matching degree between the initial object and the industry, thereby solving the technical problem of determining the first industry matching degree between the initial object and the industry.

[0094] Embodiment 4:

[0095] The second industry matching degree determination unit 2022 is further introduced.

[0096] The object evaluation information includes subjective evaluation information and objective evaluation information.

[0097] The second industry matching degree determining unit 2022 is used to determine the subjective matching degree between the initial object and the industry for any initial object among the multiple initial objects based on the subjective evaluation information of the initial object and the industry development trend information, wherein the subjective evaluation information is the evaluation information submitted by the initial object. The objective matching degree between the initial object and the industry is determined based on the objective evaluation information of the initial object and the industry development trend information, wherein the objective evaluation information is the evaluation information submitted by the target object recommendation platform that recommended the initial object. The second industry matching degree between the initial object and the industry is determined based on the subjective matching degree and the objective matching degree between the initial object and the industry.

[0098] In a possible implementation, for any of the multiple initial objects, the subjective matching degree between the initial object and the industry is determined based on the subjective evaluation information of the initial object and the industry development trend information. The subjective evaluation information is the evaluation information submitted by the initial object. This step can be implemented in a variety of ways. For example, natural language processing technology can be used to analyze keywords in the subjective evaluation information, and then match it with the industry development trend information. Another way is to establish a scoring model to give scores according to different dimensions in the subjective evaluation information, and then perform weighted calculations with the industry development trend information.

[0099] In one possible implementation, the objective matching degree between the initial object and the industry is determined based on the objective evaluation information of the initial object and the industry development trend information. The objective evaluation information is the evaluation information submitted by the target object recommendation platform that recommends the initial object. This step can adopt a method similar to the subjective matching degree calculation, but it is necessary to consider that the objective evaluation information may contain more structured data, such as ratings, rankings, etc. Therefore, a more complex algorithm can be designed to process these data. For example, a machine learning algorithm can be used to train a model that can predict the matching degree based on the objective evaluation information and industry development trend information.

[0100] In one possible implementation, the second industry matching degree between the initial object and the industry is determined based on the subjective matching degree and the objective matching degree between the initial object and the industry. This step can be implemented in a variety of ways. For example, the average of the subjective matching degree and the objective matching degree can be simply taken. Another more complex method is to assign different weights to the subjective matching degree and the objective matching degree according to the characteristics of different industries, and then perform weighted averaging. More complex fusion algorithms, such as support vector machines (SVM) or random forests, can also be used to comprehensively consider the subjective matching degree and the objective matching degree to obtain the final second industry matching degree.

[0101] Through this method, the application can more comprehensively and accurately evaluate the degree of match between the initial object and the industry. The subjective evaluation information reflects the initial object's own understanding and expectations of the industry, while the objective evaluation information provides a third-party professional evaluation. Combined with industry development trend information, not only the current matching status is considered, but also the future development potential. This multi-dimensional evaluation method can greatly improve the accuracy and comprehensiveness of the match.

[0102] The technical solution in Example 4 realizes a multi-dimensional evaluation of the degree of match between the initial object and the industry by introducing subjective evaluation information and objective evaluation information, and combining it with industry development trend information. First, the solution calculates the subjective matching degree and the objective matching degree separately, so that the evaluation of the initial object itself and the evaluation of the third party can be fully considered. Secondly, the industry development trend information is introduced in the calculation process, so that the matching evaluation not only considers the current situation, but also considers future development. Finally, by comprehensively considering the subjective and objective matching degrees, the final second industry matching degree is obtained, achieving a more accurate and comprehensive matching evaluation.

[0103] As a preferred implementation, the following specific steps can be used to implement the above technical solution:

[0104] 1. Data collection: Collect the subjective and objective evaluation information of the initial object. Subjective evaluation information can be collected through questionnaires or online forms, including the initial object's evaluation of his or her own ability, interests, career planning, etc. Objective evaluation information can be obtained from the target object recommendation platform, including the initial object's skill score, work performance evaluation, etc. At the same time, collect industry development trend information, which can be obtained from industry reports, expert forecasts and other channels.

[0105] 2. Data preprocessing: Clean and standardize the collected data. For example, convert text data into structured data and unify the scoring standards from different sources.

[0106] 3. Subjective matching calculation: Use natural language processing technology to analyze subjective evaluation information and extract keywords and topics. Then, use algorithms such as cosine similarity to calculate the similarity between these keywords and topics and industry development trend information to obtain a subjective matching score.

[0107] 4. Objective matching calculation: Use a machine learning algorithm (such as random forest or gradient boosting tree) to train a model, with objective evaluation information and industry development trend information as input and objective matching score as output.

[0108] 5. Matching degree fusion: Use the weighted average method to fuse subjective matching degree and objective matching degree. The weight can be adjusted according to the characteristics of different industries. For example, for industries that require innovation capabilities, a higher weight can be given to subjective matching degree; for industries that require stability, a higher weight can be given to objective matching degree.

[0109] 6. Result output: The calculated second industry matching degree is output as the final result. A threshold value can be set, and the initial objects above the threshold value are considered to have a high matching degree with the industry.

[0110] Through this implementation, the technical solution in Example 4 can make full use of both subjective and objective evaluation information, combined with industry development trends, to obtain more comprehensive and accurate matching results. This method not only takes into account the self-evaluation of the initial object, but also the professional evaluation of the third party, while also taking into account the future development of the industry, thereby greatly improving the accuracy and foresight of the matching.

[0111] Embodiment 5:

[0112] The third industry matching degree determination unit 2023 is further introduced.

[0113] The education information includes academic degree information and skill training information, and the evaluation information includes subjective evaluation information and objective evaluation information.

[0114] The third industry matching degree determining unit 2023 is used to determine, for any initial object among the multiple initial objects, a first evaluation matching degree between the initial object and the industry based on the skill training information of the initial object and the industry evaluation information. Determine a second evaluation matching degree between the initial object and the industry based on the subjective evaluation information, objective evaluation information and the industry evaluation information of the initial object. Determine a third industry matching degree between the initial object and the industry based on the skill evaluation matching degree and the evaluation matching degree between the initial object and the industry.

[0115] In a possible implementation, for any of the multiple initial objects, based on the skill training information of the initial object and the industry evaluation information, the first evaluation matching degree between the initial object and the industry is determined. This step can be implemented in a variety of ways. For example, natural language processing technology can be used to analyze keywords in the skill training information and then match them with the industry evaluation information. Another way is to establish a scoring model to give scores according to different dimensions in the skill training information, and then perform weighted calculations with the industry evaluation information.

[0116] In a possible implementation, the second evaluation matching degree between the initial object and the industry is determined based on the subjective evaluation information, objective evaluation information and the industry evaluation information of the initial object. This step can adopt a method similar to the subjective matching degree calculation, but it is necessary to consider that the subjective evaluation information and the objective evaluation information may contain more structured data. Therefore, a more complex algorithm can be designed to process these data. For example, a machine learning algorithm can be used to train a model that can predict the second evaluation matching degree based on the subjective evaluation information, the objective evaluation information and the industry evaluation information.

[0117] In one possible implementation, the third industry matching degree between the initial object and the industry is determined based on the skill evaluation matching degree and the evaluation matching degree between the initial object and the industry. This step can be implemented in a variety of ways. For example, the average of the skill evaluation matching degree and the evaluation matching degree can be simply taken. Another more complex method is to assign different weights to the skill evaluation matching degree and the evaluation matching degree according to the characteristics of different industries, and then perform weighted averaging. More complex fusion algorithms, such as support vector machines (SVM) or random forests, can also be used to comprehensively consider the skill evaluation matching degree and the evaluation matching degree to obtain the final second industry matching degree.

[0118] Embodiment 6:

[0119] The first fusion unit 2024 is further introduced.

[0120] In a possible implementation, the first fusion unit 2024 is used to weightedly fuse the first industry matching degree, the second industry matching degree, and the third industry matching degree between each initial object and the industry to obtain the matching degree between each initial object and the industry.

[0121] Among them, the weights of weighted fusion are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.

[0122] Embodiment 7:

[0123] The candidate object determination module 203 of the system is introduced. The enterprise information includes enterprise culture information, enterprise style information and enterprise planning information. The enterprise evaluation information includes work intensity evaluation information, work content evaluation information and work development evaluation information. The object information includes historical work information and object evaluation information. Figure 4 The candidate object determination module 203 includes:

[0124] The first enterprise matching degree determining unit 2031 is used to determine the first enterprise matching degree between each reference object and the target enterprise based on the enterprise culture information, the work intensity evaluation information, and the historical work information and object evaluation information of each reference object.

[0125] The second enterprise matching degree determining unit 2032 is used to determine the second enterprise matching degree between each reference object and the target enterprise based on the enterprise style information, the work content evaluation information and the object evaluation information of each reference object.

[0126] The third enterprise matching degree determining unit 2033 is used to determine the third enterprise matching degree between each reference object and the target enterprise based on the enterprise planning information, the work development evaluation information and the object evaluation information.

[0127] The second fusion unit 2034 is used to fuse the first enterprise matching degree, the second enterprise matching degree and the third enterprise matching degree between each reference object and the enterprise to obtain the matching degree between each reference object and the target enterprise.

[0128] The second matching unit 2035 is used to determine the reference objects among the multiple reference objects whose matching degree with the enterprise is greater than or equal to the second matching degree threshold as candidate objects, so as to obtain the multiple candidate objects.

[0129] Among them, enterprise information includes enterprise culture information, enterprise style information and enterprise planning information. These three types of information can describe the characteristics of the enterprise from multiple perspectives. For example, enterprise culture information can include the enterprise's values, mission and vision; enterprise style information can include management style, work atmosphere and team collaboration methods; enterprise planning information can include the enterprise's development strategy and future plans. Enterprise evaluation information includes work intensity evaluation information, work content evaluation information and work development evaluation information. These evaluation information can come from internal employees, former employees or third-party evaluation agencies. Work intensity evaluation information can include working hours, stress level, etc.; work content evaluation information can include the challenge and innovation of the work, etc.; work development evaluation information can include promotion opportunities, training opportunities, etc. Object information includes historical work information and object evaluation information. Historical work information can include past work experience, position, industry, etc.; object evaluation information can include evaluations from former employers and colleagues, etc.

[0130] The technical solution of this application determines the candidate objects through the following steps:

[0131] First, based on the corporate culture information, work intensity evaluation information, and the historical work information and object evaluation information of each reference object, the first corporate matching degree between each reference object and the target enterprise is determined. This step mainly considers whether the reference object is consistent with the corporate culture and whether it can adapt to the work intensity of the enterprise.

[0132] Secondly, based on the enterprise style information, work content evaluation information and object evaluation information of each reference object, the second enterprise matching degree between each reference object and the target enterprise is determined. This step mainly considers whether the reference object is adapted to the work style of the enterprise and whether he is interested in the work content.

[0133] Thirdly, based on the enterprise planning information, work development evaluation information and object evaluation information, the third enterprise matching degree between each reference object and the target enterprise is determined. This step mainly considers whether the career development direction of the reference object is consistent with the enterprise's plan.

[0134] Then, the first enterprise matching degree, the second enterprise matching degree and the third enterprise matching degree between each reference object and the enterprise are fused to obtain the matching degree between each reference object and the target enterprise. The fusion method can adopt weighted average, geometric average or other more complex algorithms to ensure that the matching degree of each dimension is reasonably considered.

[0135] Finally, the reference objects whose matching degree with the enterprise is greater than or equal to the second matching degree threshold are determined as candidate objects to obtain multiple candidate objects. The second matching degree threshold can be adjusted according to actual needs to control the number and quality of candidate objects.

[0136] This multi-dimensional matching method can more comprehensively evaluate the degree of match between the reference object and the target company. For example, a reference object may be a good match with the company in terms of job content, but there are big differences in corporate culture. By comprehensively considering multiple dimensions, a more balanced and comprehensive matching result can be obtained.

[0137] In addition, the technical solution of the present application is also flexible and scalable. The weights of various dimensions can be adjusted or new matching dimensions can be added according to the characteristics of different industries or enterprises. For example, for innovative enterprises, more attention may be paid to the innovation ability and learning ability of the reference object; for service-oriented enterprises, more attention may be paid to the communication ability and service awareness of the reference object.

[0138] As a preferred implementation, a machine learning algorithm can be used to optimize the matching process. By analyzing historical successful matching cases, the system can continuously learn and adjust the weights and matching rules of each dimension, thereby improving the accuracy of matching. For example, algorithms such as support vector machines (SVM) or random forests can be used to train the matching model, with input features including enterprise information, enterprise evaluation information, and object information of reference objects, and the output is a matching degree score.

[0139] In practical applications, the technical solution of this application can be seamlessly integrated with other recruitment processes. For example, after initially screening out candidates, you can further combine interview evaluation, skill test and other information to conduct a more in-depth assessment of the candidates. This multi-stage, multi-dimensional assessment method can greatly improve the accuracy and efficiency of talent matching.

[0140] The technical solution provided in Example 7 further proposes a technical solution for determining multiple candidate objects from multiple reference objects based on the enterprise information, enterprise evaluation information of the target enterprise and object information of multiple reference objects.

[0141] The technical solution provided in Example 7 uses multi-dimensional and multi-level matching calculations to comprehensively evaluate the matching degree between the reference object and the target enterprise from multiple perspectives such as corporate culture, work intensity, corporate style, work content, corporate planning and work development. This method not only takes into account the objective information of the enterprise, but also combines the evaluation information, while also making full use of the historical work information and evaluation information of the object, so as to more accurately identify the candidate objects with a high matching degree with the target enterprise.

[0142] Compared with traditional simple keyword matching or single-dimensional screening, this multi-dimensional matching method can more comprehensively and deeply evaluate the matching degree between the reference object and the target enterprise, effectively improving the accuracy and efficiency of matching. At the same time, by setting the matching degree threshold, the quantity and quality of candidate objects can be flexibly controlled, laying the foundation for subsequent accurate recommendations.

[0143] Embodiment 8:

[0144] The first enterprise matching degree determination unit 2031 is further introduced.

[0145] The first enterprise matching degree determining unit 2031 is used to determine the enterprise culture matching degree between the reference object and the target enterprise for any reference object among the multiple reference objects based on the enterprise culture information and the object evaluation information of the reference object. The work intensity matching degree between the reference object and the target enterprise is determined based on the work intensity evaluation information and the historical work information of the reference object. The first enterprise matching degree between the reference object and the target enterprise is determined based on the enterprise culture matching degree and the work intensity matching degree between the reference object and the target enterprise.

[0146] Among them, the determination of the degree of corporate culture matching can be achieved in a variety of ways. For example, natural language processing technology can be used to analyze corporate culture information and object evaluation information of reference objects, extract keywords and calculate semantic similarity. Another method is to establish corporate culture feature vectors and reference object feature vectors, and evaluate the degree of matching by calculating vector distance.

[0147] There are also many ways to determine the degree of work intensity matching. You can analyze specific indicators in the work intensity evaluation information, such as working hours, task complexity, etc., and compare them with the historical work information of the reference object. Another method is to use a machine learning model to train a prediction model based on historical data to evaluate whether the reference object can adapt to the work intensity of the target enterprise.

[0148] The combination of corporate culture matching and work intensity matching can be achieved through weighted average, multi-objective optimization or machine learning models. This combination not only takes into account the importance of the two factors, but also adjusts the weight of each factor according to the needs of different companies.

[0149] The technical solution provided in Example 8 solves the problem of how to determine the first enterprise matching degree between the reference object and the target enterprise by first analyzing the enterprise culture information and the object evaluation information of the reference object to determine the enterprise culture matching degree. This step can effectively evaluate whether the reference object is suitable for the cultural environment of the target enterprise, thereby predicting its adaptability and potential performance in the enterprise.

[0150] Next, the technical solution provided in Example 8 uses the work intensity evaluation information and the historical work information of the reference object to determine the degree of work intensity matching. This step can evaluate whether the reference object can adapt to the work rhythm and pressure of the target enterprise, thereby predicting its performance in actual work.

[0151] Finally, the technical solution provided in Example 8 combines the corporate culture matching degree and the work intensity matching degree to obtain the first corporate matching degree. This comprehensive evaluation method not only considers the skills and experience of the reference object, but also considers its adaptability in terms of corporate culture and work intensity, thereby providing a more comprehensive and accurate matching result.

[0152] Through this multi-dimensional evaluation method, this application can more accurately identify candidates suitable for the target company, improve recruitment efficiency, and reduce the risk of talent loss. At the same time, it can also help job seekers find a more suitable working environment and improve employment satisfaction.

[0153] Example 8 further proposes that for any reference object among multiple reference objects, based on the corporate culture information and the object evaluation information of the reference object, the degree of corporate culture matching between the reference object and the target enterprise is determined; based on the work intensity evaluation information and the historical work information of the reference object, the degree of work intensity matching between the reference object and the target enterprise is determined; based on the corporate culture matching degree and the work intensity matching degree between the reference object and the target enterprise, the first enterprise matching degree between the reference object and the target enterprise is determined.

[0154] The technical solution provided in Example 8 improves the accuracy and comprehensiveness of matching by considering two important factors: corporate culture and work intensity. It not only focuses on skill matching, but also considers cultural adaptability and work intensity adaptability, which are very important for ensuring long-term employment relationships and employee satisfaction.

[0155] The innovation of the technical solution provided in Example 8 is to combine qualitative corporate culture information with quantitative work intensity evaluation to form a comprehensive matching evaluation system. This method is more comprehensive and effective than the traditional matching method based only on skills and experience, and can better predict the performance and adaptability of job seekers in a specific corporate environment.

[0156] As a preferred implementation, the technical solution provided in Example 8 can be implemented by the following steps:

[0157] 1. Calculation of corporate culture matching: Use natural language processing technology to analyze corporate culture information and extract keywords such as "innovation", "teamwork", "customer orientation", etc. Similarly, analyze the object evaluation information of the reference object and extract relevant keywords. Calculate the cosine similarity of the two sets of keywords and get a score between 0 and 1, which represents the corporate culture matching.

[0158] 2. Work intensity matching calculation: Analyze the work intensity evaluation information and extract specific indicators such as average working hours, overtime frequency, task complexity, etc. At the same time, analyze the historical work information of the reference object and extract the corresponding indicators. Normalize each indicator, and then calculate the Euclidean distance to obtain a score between 0 and 1, which represents the work intensity matching degree.

[0159] 3. Calculation of the first enterprise matching degree: Set weights w1 and w2 (w1+w2=1), which represent the importance of enterprise culture matching and work intensity matching respectively. Use the weighted average method to calculate the final first enterprise matching degree: First enterprise matching degree = w1*enterprise culture matching degree + w2*work intensity matching degree

[0160] In this way, a comprehensive matching score can be generated for each reference object, thereby helping the target company to better screen suitable candidates.

[0161] Embodiment 9:

[0162] The second enterprise matching degree determination unit 2032 is further introduced.

[0163] The object evaluation information includes subjective evaluation information and objective evaluation information.

[0164] The second enterprise matching degree determining unit 2032 is used to determine the enterprise style matching degree between the reference object and the target enterprise for any reference object among the multiple reference objects based on the enterprise style information and the objective evaluation information of the reference object, the objective evaluation information being the evaluation information submitted by the target object recommendation platform that recommended the initial object. The work content matching degree between the reference object and the target enterprise is determined based on the work content evaluation information and the subjective evaluation information of the reference object, the subjective evaluation information being the evaluation information submitted by the initial object. The enterprise style matching degree and the work content matching degree between the reference object and the target enterprise are used to determine the second enterprise matching degree between the reference object and the target enterprise.

[0165] Among them, for any reference object among the multiple reference objects, the degree of corporate style matching between the reference object and the target enterprise is determined based on the corporate style information and the objective evaluation information of the reference object. The objective evaluation information is submitted by the target object recommendation platform that recommends the initial object. This method can reduce subjective bias and improve the objectivity of the matching. The degree of work content matching between the reference object and the target enterprise is determined based on the work content evaluation information and the subjective evaluation information of the reference object. The subjective evaluation information is submitted by the initial object. This method can fully consider the individual's feelings and preferences for the work content and improve the personalization of the matching. The degree of corporate style matching and the degree of work content matching between the reference object and the target enterprise are comprehensively considered to determine the second corporate matching degree between the reference object and the target enterprise. This comprehensive consideration method can balance objective and subjective factors to obtain more comprehensive and accurate matching results.

[0166] Specifically, the calculation of the degree of corporate style matching can adopt a multi-dimensional evaluation method. For example, the corporate style information can be decomposed into multiple dimensions such as corporate culture, management style, and work environment, and then each dimension can be evaluated separately. For each dimension, a series of evaluation indicators can be set, and corresponding scores can be given based on objective evaluation information. Finally, the overall corporate style matching degree is obtained through weighted average or other comprehensive calculation methods.

[0167] The calculation of the degree of job content matching can be done by semantic analysis and similarity calculation. First, semantic analysis is performed on the job content evaluation information and the subjective evaluation information of the reference object to extract keywords and topics. Then, the similarity between the two is calculated using a vector space model or other similarity calculation methods. In addition, the introduction of a machine learning algorithm can be considered to continuously optimize the matching model through training data.

[0168] When determining the second enterprise matching degree, a weighted average method can be used. For example, the weight of the enterprise style matching degree can be set to 0.6, and the weight of the work content matching degree can be set to 0.4. By adjusting these weights, the influence of objective factors and subjective factors on the final matching results can be flexibly balanced.

[0169] As a preferred implementation, the technical solution provided in Example 9 may also introduce a time decay factor to better reflect the timeliness of the evaluation information. For example, a smaller weight may be assigned to earlier evaluation information, while a larger weight may be assigned to newer evaluation information. This ensures that the matching results are closer to the current status of the reference object and the target enterprise.

[0170] In addition, the technical solution provided in Example 9 can also filter the matching results by setting a threshold. For example, a minimum matching degree threshold can be set, and only when the matching degree of the second enterprise exceeds this threshold, the reference object is included in the candidate object pool. This method can effectively improve the quality of recommendations and reduce irrelevant matching results.

[0171] This technical solution improves the accuracy and comprehensiveness of matching by dividing the object evaluation information into subjective evaluation information and objective evaluation information, and using them for different aspects of matching calculation. The solution first determines the matching degree of corporate style based on corporate style information and objective evaluation information, and then determines the matching degree of work content based on work content evaluation information and subjective evaluation information, and combines these two aspects to obtain the second corporate matching degree. This multi-dimensional and multi-level matching method can more comprehensively evaluate the matching degree between the reference object and the target enterprise, and effectively solves the technical problem of how to more accurately determine the second corporate matching degree between the reference object and the target enterprise.

[0172] Embodiment 10:

[0173] The third enterprise matching degree determination unit 2033 is further introduced.

[0174] The object evaluation information includes subjective evaluation information and objective evaluation information. The third enterprise matching degree determination unit 2033 is used to determine the enterprise planning matching degree between the reference object and the target enterprise for any reference object among the multiple reference objects based on the enterprise planning information and the objective evaluation information of the reference object. The work development matching degree between the reference object and the target enterprise is determined based on the work development evaluation information and the subjective evaluation information of the reference object. The third enterprise matching degree between the reference object and the target enterprise is determined based on the enterprise planning matching degree and the work development matching degree between the reference object and the target enterprise.

[0175] In a possible implementation, for any reference object among the multiple reference objects, the degree of matching of the enterprise plan between the reference object and the target enterprise is determined based on the enterprise planning information and the objective evaluation information of the reference object. The calculation of the degree of matching of enterprise plans can adopt the method of semantic analysis and similarity calculation. First, semantic analysis is performed on the enterprise planning information and the objective evaluation information of the reference object to extract keywords and topics. Then, the similarity between the two is calculated using a vector space model or other similarity calculation method. In addition, it is also possible to consider introducing a machine learning algorithm to continuously optimize the matching model through training data.

[0176] In a possible implementation, the degree of job development matching between the reference object and the target enterprise is determined based on the job development evaluation information and the subjective evaluation information of the reference object. The calculation of the job development matching degree can adopt the method of semantic analysis and similarity calculation. First, semantic analysis is performed on the job development evaluation information and the subjective evaluation information of the reference object to extract keywords and topics. Then, the similarity between the two is calculated using a vector space model or other similarity calculation methods. In addition, it is also possible to consider introducing a machine learning algorithm to continuously optimize the matching model through training data.

[0177] In a possible implementation, the third enterprise matching degree between the reference object and the target enterprise is determined based on the enterprise planning matching degree and the work development matching degree between the reference object and the target enterprise. When determining the third enterprise matching degree, a weighted average method can be used. For example, the weight of the enterprise planning matching degree can be set to 0.6, and the weight of the work development matching degree can be set to 0.4. By adjusting these weights, the influence of various factors on the final matching result can be flexibly adjusted.

[0178] Embodiment 11:

[0179] The second fusion unit 2034 is further introduced.

[0180] In a possible implementation, the second fusion unit 2034 is used to perform weighted fusion on the first enterprise matching degree, the second enterprise matching degree and the third enterprise matching degree between each reference object and the enterprise to obtain the matching degree between each reference object and the target enterprise.

[0181] Among them, the weights of weighted fusion are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.

[0182] Embodiment 12:

[0183] The target object determination module 204 of the system is introduced. The position information includes position requirement information, position description information and position development information. The object information includes historical work information, education information and object evaluation information. Figure 5 The target object determination module 204 includes:

[0184] The first position matching degree determining unit 2041 is used to determine the first position matching degree between each candidate and the target position based on the historical work information, education information and position requirement information of each candidate.

[0185] The second position matching degree determining unit 2042 is used to determine the second position matching degree between each candidate and the target position based on the historical work information of each candidate and the position description information.

[0186] The third position matching degree determining unit 2043 is used to determine the third position matching degree between each candidate object and the target position based on the object evaluation information of each candidate object and the position development information.

[0187] The third fusion unit 2044 is used to fuse the first position matching degree, the second position matching degree and the third position matching degree between each candidate object and the target position to obtain the matching degree between each candidate object and the target position.

[0188] The third matching unit 2045 is configured to determine, as a target object, a candidate object among the multiple candidate objects whose matching degree with the target position is greater than or equal to the third matching degree threshold, to obtain the at least one target object.

[0189] As a preferred implementation, the technical solution provided in Example 12 can be implemented through the following specific steps:

[0190] 1. Obtain the job information of the target position, including job requirements, job description and job development information. For example, for a software engineer position, the job requirements may include "bachelor's degree or above, more than 3 years of relevant work experience, proficient in Java programming"; the job description may include "responsible for the development and maintenance of the company's core business systems"; the job development information may include "the opportunity to participate in large-scale projects and be promoted to technical experts or management positions".

[0191] 2. Obtain object information of multiple candidate objects, including historical work information, education information, and object evaluation information. For example, for candidate A, historical work information may be "5 years of Java development experience, participated in e-commerce platform development"; education information may be "Bachelor's degree in computer science"; object evaluation information may include "strong technical ability, good at teamwork".

[0192] 3. Calculate the matching degree of the first position:

[0193] – Compare Candidate A’s education (Bachelor of Computer Science) with the job requirements (bachelor’s degree or above) to determine if the education matches.

[0194] – Compare work experience (5 years) with requirements (3+ years) to determine experience match.

[0195] – Compare skills (Java development experience) with requirements (proficiency in Java programming) to determine skill match.

[0196] – Combining the above three points, the matching degree of the first position is obtained, for example, 90%.

[0197] 4. Calculate the matching degree of the second position:

[0198] – Compare Candidate A’s previous work content (e-commerce platform development) with the job description (core business system development and maintenance) to assess relevance.

[0199] – Consider the match between the job field and the applicant.

[0200] – Comprehensively evaluate and determine the degree of match for the second position, e.g. 85%.

[0201] 5. Calculate the matching degree of the third position:

[0202] – Analyze the degree of match between candidate A’s object evaluation information (strong technical ability and good at teamwork) and the position development information.

[0203] – Evaluate the candidate’s potential for advancement.

[0204] – Comprehensively evaluate and determine the matching degree of the third position, for example 80%.

[0205] 6. Integrate three levels of matching:

[0206] – A weighted average approach can be adopted, for example, giving weights of 0.4, 0.3, and 0.3 to the first, second, and third positions respectively.

[0207] – Calculate the final matching degree: 0.4*90%+0.3*85%+0.3*80%=85.5%

[0208] 7. Set a matching threshold, such as 80%. If the final matching degree of candidate A is 85.5% higher than the threshold, it is determined as the target object.

[0209] 8. Repeat the above steps for all candidate objects, and finally obtain all target objects with matching degrees higher than the threshold.

[0210] The technical solution provided in Example 12 achieves a comprehensive evaluation of the degree of match between the candidate and the target position through multi-dimensional and multi-level matching analysis. Specifically, the position information includes position requirement information, position description information and position development information, which are used to comprehensively describe the target position. The object information includes historical work information, education information and object evaluation information, which are used to comprehensively describe the candidate.

[0211] The technical solution provided in Example 12 first matches the candidate's historical work information and education information with the position requirement information to evaluate whether the candidate meets the basic requirements of the position and obtain the first position matching degree. Secondly, based on the candidate's historical work information and the position description information, it is matched to evaluate whether the candidate's work experience is consistent with the position description to obtain the second position matching degree. Thirdly, based on the candidate's evaluation information and the position development information, it is matched to evaluate whether the candidate's potential meets the position development requirements to obtain the third position matching degree. Finally, the matching degrees of these three dimensions are integrated to obtain the overall matching degree, and the final target object is screened out by setting a threshold.

[0212] Through this multi-dimensional and multi-level matching analysis method, the technical solution provided in Example 12 not only considers whether the candidate meets the basic requirements of the position, but also evaluates the relevance of the work content and the potential for future development, so as to more comprehensively and accurately identify the candidate who is highly matched with the target position. Compared with the traditional simple keyword matching or single-dimensional screening, this method can more effectively solve the technical problem of determining the target object with a high degree of match with the target position from multiple candidates.

[0213] Embodiment 13:

[0214] The first position matching degree determining unit 2041 is further introduced.

[0215] The historical work information includes historical work content and historical positions.

[0216] The first position matching degree determining unit 2041 is used to determine, for any candidate among the multiple candidates, the content requirement matching degree between the candidate and the target position based on the candidate's historical work content and the position requirement information. The position matching degree between the candidate and the target position is determined based on the candidate's historical position and the position requirement information. The first position matching degree between the candidate and the target position is determined based on the content requirement matching degree and the position matching degree between the candidate and the target position.

[0217] Among them, by comparing the candidate's historical work content and job requirements information, the degree of content requirement matching is determined. This step can be achieved in many ways, such as using natural language processing technology to perform semantic analysis on historical work content and job requirements and calculate similarity scores. Another method is to extract keywords and perform weighted matching, and give scores based on the degree of matching.

[0218] When determining the degree of job matching, you can use job level comparison or job similarity analysis. Job level comparison can establish a job level system to compare the differences in levels between the candidate's historical job and the target job. Job similarity analysis can use a machine learning model to train the similarity relationship between different jobs based on a large amount of historical data.

[0219] The results of the content requirement matching degree and the position matching degree can be combined through weighted average or more complex fusion algorithm to obtain the first position matching degree. The weight can be adjusted according to the characteristics of different industries or positions to meet the needs of different scenarios.

[0220] This multi-dimensional matching method is more comprehensive and accurate than simple keyword matching. By considering both job content and position, it is possible to more accurately assess the match between a candidate and a target position. For example, a candidate may not have the exact same job title, but his / her job content is highly relevant to the target position, in which case this method can still identify his / her potential match.

[0221] In specific implementation, a threshold can be set, such as 0.7 (full score is 1). When the calculated first position matching degree exceeds this threshold, it is considered that the candidate has a high matching degree with the target position. The system can further sort the candidates according to this matching degree to provide more accurate talent recommendations for enterprises.

[0222] In practical applications, machine learning technology can be combined to continuously optimize the matching algorithm. By collecting data on successful recruitment cases, the system can learn which factors are more important for successful matching, thereby dynamically adjusting the weights of each dimension and improving the accuracy of matching.

[0223] Compared with the prior art, the technical solution provided in Example 13 has obvious advantages in the following aspects: First, it takes into account the two dimensions of work content and position, rather than relying solely on the matching of position titles, which greatly improves the accuracy and comprehensiveness of matching. Second, by introducing historical work information and education information, this solution can more comprehensively evaluate the background and ability of candidates, rather than relying solely on keywords in the resume. Finally, the flexibility of this solution enables it to adapt to the special needs of different industries and positions, and optimize the matching results by adjusting weights and matching algorithms.

[0224] Through this method, the technical solution provided in Example 13 can better solve the technical problem of determining the first position matching degree between the candidate and the target position, improve the accuracy of talent matching, and provide more valuable recommendation services for enterprises and job seekers.

[0225] Embodiment 14:

[0226] The second position matching degree determining unit 2042 is further introduced.

[0227] The historical work information includes historical work content and historical work industry.

[0228] The second position matching degree determining unit 2042 is used to determine, for any candidate among the multiple candidates, the content description matching degree between the candidate and the target position based on the candidate's historical work content and the position description information. The industry description matching degree between the candidate and the target position is determined based on the candidate's historical work industry and the position description information. The second position matching degree between the candidate and the target position is determined based on the content description matching degree and the industry description matching degree between the candidate and the target position.

[0229] The above method of determining the matching degree of the second position is described below.

[0230] First, determine the degree of content description matching based on the candidate's historical work content and job description information. Natural language processing techniques, such as semantic similarity analysis, can be used to compare historical work content with job descriptions. For example, a word vector model can be used to convert text into vectors, and then the cosine similarity between vectors can be calculated. The higher the similarity, the higher the degree of content description matching.

[0231] Secondly, determine the degree of industry description match based on the candidate's historical work industry and job description information. An industry classification system can be established to match the historical work industry with the industry of the target position. If they belong to the same industry or related industries, a higher match score will be given. The correlation between industries can also be considered. For example, the IT industry and the electronics industry may have a high correlation.

[0232] Finally, the content description matching degree and the industry description matching degree are weighted and merged to obtain the second position matching degree. The weights of the two can be adjusted according to actual needs. For example, the weight of the content description matching degree can be set to 0.6, and the weight of the industry description matching degree can be set to 0.4.

[0233] The technical solution provided in Example 14 evaluates the degree of match between the candidate and the target position through multiple dimensions to improve the accuracy and comprehensiveness of the match. Specifically, the technical solution provided in Example 14 evaluates the match from the aspects of content description matching, industry description matching, subjective position development matching, and objective position development matching. By comprehensively considering these factors, the technical solution provided in Example 14 can more comprehensively and accurately evaluate the degree of match between the candidate and the target position, thereby improving the accuracy and efficiency of recruitment recommendations. This multi-dimensional matching method effectively solves the limitations of traditional simple keyword matching or single-dimensional screening, can better explore the potential of job seekers and the needs of positions, and improve the quality of person-job matching.

[0234] Embodiment 15:

[0235] The third position matching degree determination unit 2043 is further introduced.

[0236] The object evaluation information includes subjective evaluation information and objective evaluation information. The third position matching degree determination unit 2043 is used to determine the subjective position development matching degree between the candidate object and the target position for any candidate object among the multiple candidate objects based on the subjective evaluation information of the candidate object and the position development information, wherein the subjective evaluation information is the evaluation information submitted by the initial object. The objective position development matching degree between the candidate object and the target position is determined based on the objective evaluation information and the position development information of the candidate object, wherein the objective evaluation information is the evaluation information submitted by the target object recommendation platform that recommended the initial object. The third position matching degree between the candidate object and the target position is determined based on the subjective position development matching degree and the objective position development matching degree between the candidate object and the target position.

[0237] The above method of determining the matching degree of the third position is explained below.

[0238] In one possible implementation, the subjective job development match is determined based on the candidate's subjective evaluation information and job development information. The subjective evaluation information may include the candidate's career plan, expected salary, work environment preference, etc. This information can be compared with the job development information. For example, if the candidate's career plan is consistent with the job development prospects, a higher match score is given.

[0239] Secondly, the objective job development matching degree is determined based on the objective evaluation information and job development information of the candidates. The objective evaluation information may include the recommendation platform's ability evaluation and performance evaluation of the candidates. These evaluations can be matched with the job requirements and development prospects. For example, if the candidate's ability evaluation results are highly consistent with the job requirements, a higher matching score will be given.

[0240] Finally, the subjective and objective job development matching degrees are weighted and integrated to obtain the third job matching degree. The weights of the two can be adjusted according to the actual situation. For example, the weight of the subjective matching degree can be set to 0.4 and the weight of the objective matching degree can be set to 0.6 to balance the influence of personal expectations and objective evaluation.

[0241] In this way, the technical solution provided in Example 15 can comprehensively consider the candidate's historical work experience, industry background, personal expectations, and objective evaluation, so as to more accurately evaluate the degree of match between the candidate and the target position. This multi-dimensional matching method not only considers the superficial job content match, but also deeply considers factors such as industry match and personal development match, which can better predict the candidate's potential performance and development possibility in the target position.

[0242] Embodiment 16:

[0243] The third fusion unit 2044 is further introduced.

[0244] In a possible implementation, the third fusion unit 2044 is used to perform weighted fusion on the first position matching degree, the second position matching degree and the third position matching degree between each candidate object and the target position to obtain the matching degree between each candidate object and the target position.

[0245] Among them, the weights of weighted fusion are set by technical personnel according to actual conditions, and the embodiments of the present application do not limit this.

[0246] Embodiment 17:

[0247] The following is an explanation of the method for determining the target object recommendation platform.

[0248] In a possible implementation, the system further includes a platform determination module, which is used to obtain platform information and platform evaluation information of multiple candidate object recommendation platforms. Based on the platform information and platform evaluation information of each candidate object recommendation platform, the credibility of each candidate object recommendation platform is determined. Among the multiple candidate object recommendation platforms, the candidate object recommendation platform whose credibility is greater than or equal to the credibility threshold is determined as the target object recommendation platform to obtain the multiple target object recommendation platforms.

[0249] Among them, the platform information is provided by each candidate object recommendation platform, and the platform evaluation information is provided by other object recommendation platforms in the multiple candidate object recommendation platforms. For example, for the first candidate object recommendation platform in the multiple candidate object recommendation platforms, the platform evaluation information of the first candidate object recommendation platform is provided by other candidate object recommendation platforms in the multiple candidate object recommendation platforms. The credibility threshold is set by the technician according to the actual situation, and the embodiment of the present application does not limit this.

[0250] All the above optional technical solutions can be arbitrarily combined to form optional embodiments of the present application, which will not be described one by one here.

[0251] The above are only optional embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. An intelligent recommendation system for recruitment and employment based on big data, characterized in that: The system comprises: an acquisition module, configured to, in response to an object recommendation request of a target enterprise, acquire object call information of the target enterprise, object information of a plurality of initial objects, industry information of the industry in which the target enterprise is located, and enterprise evaluation information of the target enterprise, wherein the plurality of initial objects are objects recommended by a plurality of target object recommendation platforms, wherein the plurality of target object recommendation platforms are object recommendation platforms whose credibility is greater than or equal to a credibility threshold, and the object call information includes enterprise information of the target enterprise and position information of a target position, wherein the target position is a position to be recommended; A reference object determination module, configured to determine a plurality of reference objects from the plurality of initial objects based on the object information of the plurality of initial objects and the industry information, wherein the reference objects are initial objects whose matching degree with the industry of the target enterprise is greater than or equal to a first matching degree threshold; a candidate object determination module, configured to determine a plurality of candidate objects from the plurality of reference objects based on the enterprise information of the target enterprise, the enterprise evaluation information of the target enterprise, and the object information of the plurality of reference objects, wherein the candidate objects are reference objects whose matching degree with the target enterprise is greater than or equal to a second matching degree threshold; a target object determination module, configured to determine at least one target object from the plurality of candidate objects based on the position information of the target position and the object information of the plurality of candidate objects, wherein the target object is a reference object whose matching degree with the target position is greater than or equal to a third matching degree threshold; A recommendation module is used to recommend the at least one target object to the target enterprise.

2. The system according to claim 1, characterized in that The object information includes historical work information, education information and object evaluation information, and the industry information includes industry description information, industry development trend information and industry evaluation information; The reference object determination module comprises: A first industry matching degree determining unit, configured to determine a first industry matching degree between each of the initial objects and the industry based on the historical work information, education information and the industry description information of each of the initial objects; A second industry matching degree determining unit, configured to determine a second industry matching degree between each of the initial objects and the industry based on the object evaluation information of each of the initial objects and the industry development trend information; A third industry matching degree determining unit, configured to determine a third industry matching degree between each of the initial objects and the industry based on the education information of each of the initial objects, the object evaluation information and the industry evaluation information; A first fusion unit is used to fuse the first industry matching degree, the second industry matching degree and the third industry matching degree between each of the initial objects and the industry to obtain the matching degree between each of the initial objects and the industry; The first matching unit is used to determine the initial objects among the multiple initial objects whose matching degree with the industry is greater than or equal to the first matching degree threshold as reference objects, so as to obtain the multiple reference objects.

3. The system according to claim 2, characterized in that The historical work information includes historical work content, historical work enterprise and historical work industry, and the education information includes historical course information, historical school information, academic degree information and skill training information; The first industry matching degree determining unit is used to determine, for any initial object among the multiple initial objects, a historical work matching degree between the initial object and the industry based on the historical work content, historical work enterprise, historical work industry and the industry description information of the initial object; Based on the academic degree information, skill training information and industry description information of the initial object, determining the degree of match between the academic degree and skills of the initial object and the industry; Using the history course information and history school information of the initial object to query, obtain the history course content and history employment direction corresponding to the initial object; based on the history course content, history employment direction and the industry description information corresponding to the initial object, determine the degree of direction matching between the initial object and the industry; Based on the historical work matching degree, educational qualification and skill matching degree, and historical employment direction between the initial object and the industry, a first industry matching degree between the initial object and the industry is determined.

4. The system according to claim 2, characterized in that The object evaluation information includes subjective evaluation information and objective evaluation information; The second industry matching degree determination unit is used to determine, for any initial object among the multiple initial objects, a subjective matching degree between the initial object and the industry based on subjective evaluation information of the initial object and the industry development trend information, wherein the subjective evaluation information is evaluation information submitted by the initial object; Determining the objective matching degree between the initial object and the industry based on the objective evaluation information of the initial object and the industry development trend information, wherein the objective evaluation information is evaluation information submitted by a target object recommendation platform that recommends the initial object; Based on the subjective matching degree and the objective matching degree between the initial object and the industry, a second industry matching degree between the initial object and the industry is determined.

5. The system according to claim 1, characterized in that The enterprise information includes enterprise culture information, enterprise style information and enterprise planning information, the enterprise evaluation information includes work intensity evaluation information, work content evaluation information and work development evaluation information, and the object information includes historical work information and object evaluation information; The candidate object determination module comprises: A first enterprise matching degree determining unit, configured to determine a first enterprise matching degree between each of the reference objects and the target enterprise based on the enterprise culture information, the work intensity evaluation information, and the historical work information and object evaluation information of each of the reference objects; A second enterprise matching degree determining unit, configured to determine a second enterprise matching degree between each of the reference objects and the target enterprise based on the enterprise style information, the work content evaluation information, and the object evaluation information of each of the reference objects; A third enterprise matching degree determining unit, configured to determine a third enterprise matching degree between each of the reference objects and the target enterprise based on the enterprise planning information, the work development evaluation information and the object evaluation information; A second fusion unit is used to fuse the first enterprise matching degree, the second enterprise matching degree and the third enterprise matching degree between each of the reference objects and the enterprise to obtain the matching degree between each of the reference objects and the target enterprise; The second matching unit is used to determine the reference objects among the multiple reference objects whose matching degree with the enterprise is greater than or equal to the second matching degree threshold as candidate objects, so as to obtain the multiple candidate objects.

6. The system according to claim 5, characterized in that The first enterprise matching degree determination unit is configured to determine, for any reference object among the plurality of reference objects, a matching degree of enterprise culture between the reference object and the target enterprise based on the enterprise culture information and the object evaluation information of the reference object; Determining a degree of matching between the work intensity of the reference object and the target enterprise based on the work intensity evaluation information and the historical work information of the reference object; Based on the corporate culture matching degree and the work intensity matching degree between the reference object and the target enterprise, a first enterprise matching degree between the reference object and the target enterprise is determined.

7. The system according to claim 5, characterized in that The object evaluation information includes subjective evaluation information and objective evaluation information; The second enterprise matching degree determination unit is used to determine, for any reference object among the multiple reference objects, the enterprise style matching degree between the reference object and the target enterprise based on the enterprise style information and objective evaluation information of the reference object, wherein the objective evaluation information is evaluation information submitted by the target object recommendation platform that recommended the initial object; Determining the degree of matching of the work content between the reference object and the target enterprise based on the work content evaluation information and the subjective evaluation information of the reference object, wherein the subjective evaluation information is the evaluation information submitted by the initial object; The second enterprise matching degree between the reference object and the target enterprise is determined based on the enterprise style matching degree and the work content matching degree between the reference object and the target enterprise.

8. The system according to claim 1, characterized in that The position information includes position requirement information, position description information and position development information, and the object information includes historical work information, education information and object evaluation information; The target object determination module includes: A first position matching degree determining unit, configured to determine a first position matching degree between each of the candidate objects and the target position based on the historical work information, education information and the position requirement information of each of the candidate objects; A second position matching degree determining unit, configured to determine a second position matching degree between each of the candidate objects and the target position based on the historical work information and the position description information of each of the candidate objects; A third position matching degree determining unit, configured to determine a third position matching degree between each of the candidate objects and the target position based on the object evaluation information and the position development information of each of the candidate objects; A third fusion unit is used to fuse the first position matching degree, the second position matching degree and the third position matching degree between each candidate object and the target position to obtain the matching degree between each candidate object and the target position; The third matching unit is used to determine, as a target object, a candidate object among the multiple candidate objects whose matching degree with the target position is greater than or equal to the third matching degree threshold, to obtain the at least one target object.

9. The system according to claim 8, characterized in that The historical work information includes historical work content and historical positions; The first position matching degree determining unit is configured to determine, for any candidate among the plurality of candidate objects, a content requirement matching degree between the candidate object and the target position based on the candidate object's historical work content and the position requirement information; Determining the degree of job matching between the candidate and the target job based on the candidate's previous job and the job requirement information; Based on the content requirement matching degree and the position matching degree between the candidate object and the target position, a first position matching degree between the candidate object and the target position is determined.

10. The system according to claim 8, characterized in that The historical work information includes historical work content and historical work industry; The second position matching degree determining unit is used to determine, for any candidate among the multiple candidate objects, a content description matching degree between the candidate object and the target position based on the historical work content of the candidate object and the position description information; Determining the degree of industry description matching between the candidate and the target position based on the candidate's historical work industry and the position description information; Based on the content description matching degree and the industry description matching degree between the candidate object and the target position, a second position matching degree between the candidate object and the target position is determined.

Images