Intelligent recruitment system and method based on LLM multi-terminal application

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an intelligent recruitment system and method based on an LLM multi-terminal application, and the system comprises a multi-terminal interaction interface module, an information interaction and processing module, a large model analysis module, a database management module, and a result display and feedback module. The multi-terminal interaction interface module provides a unified user access point at an App terminal, a WeChat applet terminal, a jitter applet terminal and a PC terminal; according to the method, the information is rapidly processed and matched, the model is trained in advance by adopting a pre-training mode based on the strong feature understanding capability of the deep neural network and the rapid parallel computing capability of the LLM large model, so that a large amount of job seeker information and post demand information can be rapidly processed in the reasoning stage; posts matched with the job seekers can be rapidly screened out from the massive post libraries, and similarly, after an enterprise owner issues recruitment requirements, the system can also rapidly find out potential appropriate personnel choices from numerous job seekers; according to the invention, on the basis of a self-trained LLM large model, the chat chat robot is adopted to interact with the enterprise owner and the job seeker through a multi-round question and answer chat mode, so that precise and intelligent recommendation of the post and the job seeker is realized, the recruitment period is greatly shortened, the enterprise can more quickly fill up the post vacancy, the business operation is smoother, and the recruitment efficiency is improved. And job seekers can obtain work opportunities more quickly, so that the job-seeking waiting time is shortened.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent recruitment system and method based on LLM multi-terminal application. Background Art

[0002] In today's era of rapid digital development, the human resources market has become more active and complex than ever before. The traditional recruitment model has gradually exposed many limitations in dealing with the problem of matching the growing demand for talent with the supply of positions.

[0003] In the traditional recruitment model, on the one hand, during the recruitment process, enterprises face a large number of resumes. Manual screening is not only time-consuming and laborious, but also often leads to low recruitment efficiency due to differences in subjective judgment of human resources specialists and the difficulty in comprehensively and accurately evaluating the compatibility between job seekers' multi-dimensional abilities and job requirements. Many potential talents are missed, and inappropriate personnel may even be introduced, increasing the company's employment costs and risks. On the other hand, job seekers often lack effective screening tools and accurate job recommendations in the vast amount of recruitment information, making it difficult to quickly locate job opportunities that are truly suitable for their own skills, career plans and expectations. They frequently wander between mismatched positions, with a long job search cycle and a low success rate. Therefore, an intelligent recruitment system and method based on LLM multi-terminal application is proposed. Summary of the invention

[0004] In view of this, the present invention provides an intelligent recruitment system and method based on LLM multi-terminal application to solve or alleviate the technical problems existing in the prior art and at least provide a beneficial option.

[0005] The technical solution of the present invention is implemented as follows: an intelligent recruitment system based on LLM multi-terminal application, including a multi-terminal interactive interface module, an information interaction and processing module, a large model analysis module, a database management module and a result display and feedback module;

[0006] The multi-terminal interactive interface module provides a unified user access point on the App side, WeChat applet side, Douyin applet side and PC side;

[0007] The information interaction and processing module monitors various input information of multiple users in real time, encrypts the collected information and transmits it to the large model analysis module, and feeds back the processing results of the large model to the corresponding interface display module, temporarily stores commonly used and hot data, and reduces the resource consumption of repeated data acquisition and transmission;

[0008] The large model analysis module selects and adapts an LLM large model suitable for the recruitment scenario in the model selection and adaptation process, and conducts in-depth optimization training on it based on professional knowledge, industry characteristics, and common expression habits in the recruitment field. The job seeker analysis and position matching function utilizes the powerful semantic understanding and analysis capabilities of the large model to deeply mine and extract features from the personal information, educational background, work experience, and skills of job seekers, and then conducts multi-dimensional and refined matching with the massive job requirements in the position library. The enterprise owner demand analysis and job seeker matching function targets the recruitment position information provided by the enterprise owner. The large model accurately interprets the educational background, age, professional skills, and work experience requirements for job seekers, and searches and locates potential suitable candidates in the job seeker database;

[0009] The database management module regularly collects position information from multiple authoritative channels and stores the data of job seekers and enterprises;

[0010] The result display and feedback module is responsible for receiving the matching result data transmitted by the large model analysis module, and formatting and visually displaying it according to the display characteristics of different terminals and user needs. For the job seeker terminal, it presents the recommended position list. For the enterprise owner terminal, it orderly displays the information of recommended job seekers, and establishes a perfect user feedback collection mechanism. After job seekers or enterprise owners view the recommended results, they can put forward specific modification opinions and suggestions.

[0011] Further preferably, the multi-terminal interaction interface module includes a unified design principle, a job seeker interface, and an employer interface. The unified design principle adopts a unified visual style and interaction design specification on the App side, WeChat mini-program side, Douyin mini-program side, and PC side. The job seeker interface: sets a prominent "Job Seeker Entrance", and after clicking, it enters the exclusive interface for job seekers. The interface mainly includes a personal information editing area, which guides job seekers to input name, age, gender, contact information, graduation school, major, education level, work experience, obtained certificates, self-evaluation, and job seeking intention information in the form of a form. At the same time, there is an intelligent chat window, which interacts with job seekers in an anthropomorphic image, collects information and answers questions. In addition, there is a job recommendation display area, which presents recommended jobs in the form of a list. Each job entry displays the job name, company name, salary range, and work location information, and provides a "View Details" button. After clicking, detailed job descriptions, job requirements, welfare benefits, etc. can be expanded. There is also an "Apply for Job" button, and job seekers can apply for their favorite jobs with one click. The employer interface enters through the "Employer Entrance" and includes an enterprise information registration and editing area, which requires employers to fill in enterprise name, enterprise type, enterprise scale, enterprise introduction, contact person's name, contact phone number, and email information. There is a recruitment job posting area where employers can post recruitment jobs. Through the intelligent chat window, they can assist in filling in job name, job responsibilities, educational requirements, major requirements, work experience requirements, skill requirements, salary treatment, number of recruits, work location, and work time arrangement information. It is equipped with an intelligent chat window for communicating with employers about recruitment needs and strategies, as well as a recommended job seeker display area, which displays information such as the name, photo, education level, major, work experience summary, and matching score of recommended job seekers. Clicking can view the complete resume, and there is a "Contact Job Seeker" button for employers to communicate further with one click.

[0012] Further preferably, the information interaction and processing module includes an information collection sub-module, an information transmission sub-module, and an information caching sub-module. The information collection sub-module is responsible for monitoring the user input information in the intelligent chat windows and information editing areas of each end, and real-time collecting the personal information and job seeking intention of job seekers, as well as the enterprise information and recruitment requirement data of employers. The information transmission sub-module encrypts the preliminarily processed information and transmits it to the large model analysis module, using a secure and reliable network transmission protocol. At the same time, it receives the matching results and analysis data returned by the large model analysis module and transmits them to the corresponding interface display module. The information caching sub-module establishes a local caching mechanism to cache common data.

[0013] Further preferably, the large model analysis module includes model selection and adaptation, job seeker analysis and job matching, and employer demand analysis and job seeker matching. The model selection and adaptation selects an LLM large model suitable for natural language processing tasks, and performs targeted fine-tuning training on the large model according to professional terms, industry knowledge, and common expression habits in the recruitment field, and constructs a dedicated recruitment field corpus, including a large number of real job descriptions, job seeker resumes, and recruitment conversation record data for model training and optimization. For the job seeker analysis and job matching, it receives the job seeker information transmitted by the information interaction and processing module, comprehensively analyzes it using the large model, extracts characteristics such as the key skills, knowledge level, and professional qualities of the job seeker, and performs multi-dimensional matching with the job requirements in the job library, using a combination of multiple algorithms such as semantic similarity calculation, keyword matching, and rule-based matching to evaluate the matching degree of the job seeker with each job. For the employer demand analysis and job seeker matching, for the recruitment information provided by the employer, the large model deeply analyzes its requirements for the job seeker's education, age, and ability, searches for candidates who meet the requirements in the job seeker database, and also uses multi-dimensional matching algorithms to evaluate the matching degree of the job seeker with the recruitment position.

[0014] Further preferably, the database management module includes a job database, a job seeker database, and an enterprise database, which are respectively used to store job information, personal information, and enterprise information.

[0015] Further preferably, the result display and feedback module is responsible for receiving the matching result data transmitted by the large model analysis module, and performing careful formatting processing and visual display according to the display characteristics of different terminals and user needs.

[0016] An intelligent recruitment method based on artificial intelligence multi-terminal applications includes the following steps:

[0017] Step 1: Job seeker recruitment or employer recruitment

[0018] Step 2: Intelligent interview.

[0019] Further preferably, in the first step, the job seeker opens the application at either end, clicks the "Job Seeker Registration" button, enters the mobile phone number or email address, obtains the verification code and fills it in to complete the registration. Then, according to the system prompts, the job seeker fills in personal basic information, educational background, work experience, and job hunting intention in sequence. During the filling process, the intelligent chat window can provide help and guidance at any time to answer the job seeker's questions about information filling. After the information filling is completed, the information interaction and processing module transmits the job seeker's information to the large model analysis module. The large model analysis module conducts a matching search in the job database based on the comprehensive situation of the job seeker, recommends suitable jobs for the job seeker, and returns the recommended job information to the job recommendation display area on the job seeker's side, arranged in descending order of matching degree. The job seeker browses the recommended job list, clicks on the interested job to view the details. If the job seeker decides to apply for the job, clicks the "Apply for Job" button, and the system automatically generates a job application and sends it to the corresponding business owner. The job seeker can also further communicate with the chatbot in the intelligent chat window about job-related issues. The business owner clicks the "Business Owner Registration" button at the corresponding end, fills in the basic information of the enterprise, and clicks the "Post Recruitment Job" button. The business owner collaborates with the chatbot in the intelligent chat window to fill in the recruitment job information. The chatbot will ask questions and confirm based on the partial information provided by the business owner. After the recruitment job information is released, the large model analysis module searches for job seekers who meet the requirements in the job seeker database, and the large model analysis module recommends the matching job seekers to the business owner. The business owner views the recommended job seeker information in the job seeker recommendation display area, and clicks on the job seeker's name to view the detailed resume. The business owner selects a suitable job seeker according to the matching degree of the job seeker and their own needs, and clicks the "Contact Job Seeker" button to arrange the subsequent interview session. In the second step, when both the business owner and the job seeker reach a preliminary intention, the business owner can initiate an intelligent interview within the system, select the interview time and notify the job seeker. The system generates an interview interface on the application sides of both parties. Before the interview starts, the large model generates a personalized interview question bank according to the job requirements and the situation of the job seeker, and loads it into the interview question display area. When the interview starts, the chatbot introduces the interview process and rules in the video call area in voice or text form. After the job seeker answers the questions, the chatbot uses speech recognition technology to convert the voice answer into text, and then uses natural language processing technology to analyze the answer content, evaluate the job seeker's professional knowledge level, language expression ability, and adaptability, and compare it with the job requirements to generate a real-time interview evaluation report, which is displayed on the interview interface on the business owner's side, including scores for various abilities, matching degree evaluation with the job, highlights and deficiencies of the answers, etc. The business owner can supplement questions or conduct further communication in the text communication area based on the feedback of the chatbot and their own judgment. After the interview ends, the business owner records the interview results in the system, and the system automatically feeds back the interview results to the job seeker, and updates the relevant information in the job seeker database and the job database according to the interview results and process data.

[0020] Due to the above technical solutions adopted in the embodiments of the present invention, it has the following advantages:

[0021] 1. The present invention enables fast information processing and matching. Based on the powerful computing ability of the large language model (LLM), the system can instantaneously process a large amount of job seeker information and job requirement information. The large model can quickly screen out jobs that match the job seekers from a vast job database. Similarly, after the business owner publishes the recruitment requirements, the system can also quickly identify potential suitable candidates from numerous job seekers, significantly shortening the recruitment cycle. Enterprises can fill job vacancies faster, enabling smoother business operations, and job seekers can also obtain job opportunities more quickly, reducing job hunting waiting time.

[0022] 2. The present invention supports multi-terminal applications on the App side, WeChat mini-program side, Douyin mini-program side, and PC side, enabling users to perform recruitment-related operations anytime and anywhere. Job seekers can complete their personal profiles, browse recommended jobs, and interact with chatbots during fragmented time; business owners can also conveniently publish recruitment information, view job seeker recommendations, and communicate. The multi-terminal synchronous and consistent user experience greatly improves the convenience and flexibility of users using the system.

[0023] The above summary is for the purpose of the specification only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present invention will be readily apparent by reference to the drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0025] Figure 1 It is the system flow chart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] In the following, only some exemplary embodiments are briefly described. As those skilled in the art can recognize, the described embodiments can be modified in various different ways without departing from the spirit or scope of the present invention. Therefore, the drawings and the description are considered to be exemplary in nature rather than restrictive.

[0027] The embodiments of the present invention will be described in detail below with reference to the drawings.

[0028] Such as Figure 1As shown in the figure, an embodiment of the present invention provides an intelligent recruitment system based on LLM multi-terminal applications, including a multi-terminal interaction interface module, an information interaction and processing module, a large model analysis module, a database management module, and a result display and feedback module;

[0029] The multi-terminal interaction interface module provides a unified user access point on the App side, WeChat mini-program side, Douyin mini-program side, and Pc side;

[0030] The information interaction and processing module monitors various input information of multi-terminal users in real time, encrypts the collected information and transmits it to the large model analysis module, and at the same time feeds back the processing results of the large model to the corresponding interface display module, and temporarily stores common and hot data to reduce the resource consumption of repeated data acquisition and transmission;

[0031] The large model analysis module selects and adapts an LLM large model suitable for the recruitment scenario in the model selection and adaptation link, and deeply optimizes and trains it according to professional knowledge, industry characteristics, and common expression habits in the recruitment field. The job seeker analysis and position matching function uses the powerful semantic understanding and analysis ability of the large model to deeply mine and extract features from the personal information, educational background, work experience, and skills of job seekers, and then performs multi-dimensional and refined matching with the massive job requirements in the position library. The enterprise owner demand analysis and job seeker matching function targets the recruitment position information provided by the enterprise owner, and the large model accurately interprets its requirements for job seekers' education, age, professional skills, and work experience, and searches and locates potential suitable candidates in the job seeker database;

[0032] The database management module regularly collects position information from multiple authoritative channels and stores the data of job seekers and enterprises;

[0033] The result display and feedback module is responsible for receiving the matching result data transmitted by the large model analysis module, formatting and visually displaying it according to the display characteristics and user needs of different terminals. For the job seeker side, it presents the recommended position list, and for the enterprise owner side, it displays the information of recommended job seekers in an orderly manner. A perfect user feedback collection mechanism is established. After job seekers or enterprise owners view the recommended results, specific modification opinions and suggestions are put forward.

[0034] In one embodiment, the multi - terminal interaction interface module includes a unified design principle, a job seeker interface, and an employer interface. The unified design principle adopts a unified visual style and interaction design specification on the App side, WeChat mini - program side, Douyin mini - program side, and PC side. The job seeker interface: Sets a prominent "Job Seeker Entrance", and after clicking, it enters the exclusive interface for job seekers. The interface mainly includes a personal information editing area, which guides job seekers to input name, age, gender, contact information, graduated school, major, educational level, work experience, obtained certificates, self - evaluation, and job - seeking intention information in the form of a form. At the same time, there is an intelligent chat window that interacts with job seekers in an anthropomorphic image, collects information and answers questions. In addition, there is a job recommendation display area that presents recommended jobs in the form of a list. Each job entry displays the job name, company name, salary range, and work location information, and provides a "View Details" button. After clicking, detailed job descriptions, job requirements, welfare benefits, etc. can be expanded. There is also an "Apply for Job" button for job seekers to apply for their desired jobs with one click. The employer interface is entered through the "Employer Entrance" and includes an enterprise information registration and editing area, which requires employers to fill in enterprise name, enterprise type, enterprise scale, enterprise introduction, contact person's name, contact phone number, and email information. There is a recruitment job posting area where employers can post recruitment jobs. With the assistance of the intelligent chat window, they can fill in job name, job responsibilities, educational requirements, major requirements, work experience requirements, skill requirements, salary treatment, number of recruits, work location, work time arrangement information. It is equipped with an intelligent chat window for communicating with employers about recruitment needs and strategies, as well as a recommended job seeker display area that shows information such as the name, photo, educational background, major, work experience summary, and matching score of recommended job seekers. Clicking on it can view the complete resume, and there is a "Contact Job Seeker" button for employers to communicate further with one click.

[0035] In one embodiment, the information interaction and processing module includes an information collection sub - module, an information transmission sub - module, and an information caching sub - module. The information collection sub - module is responsible for monitoring the user - input information in the intelligent chat windows and information editing areas of each terminal, and real - time collecting the personal information and job - seeking intentions of job seekers, as well as the enterprise information and recruitment requirement data of employers. The information transmission sub - module encrypts the preliminarily processed information and transmits it to the large - model analysis module, using a secure and reliable network transmission protocol. At the same time, it receives the matching results and analysis data returned by the large - model analysis module and transmits them to the corresponding interface display module. The information caching sub - module establishes a local caching mechanism to cache commonly used data.

[0036] In one embodiment, the large model analysis module includes model selection and adaptation, job seeker analysis and job matching, and employer demand analysis and job seeker matching. The model selection and adaptation selects an LLM large model suitable for natural language processing tasks, and performs targeted fine-tuning training on the large model according to the professional terms, industry knowledge, and common expression habits in the recruitment field, and constructs a dedicated recruitment field corpus, including a large number of real job descriptions, job seeker resumes, and recruitment conversation record data, for model training and optimization. For job seeker analysis and job matching, it receives the job seeker information transmitted by the information interaction and processing module, comprehensively analyzes it using the large model, extracts features such as the key skills, knowledge level, and professional qualities of the job seeker, and performs multi-dimensional matching with the job requirements in the job library, using a combination of various algorithms such as semantic similarity calculation, keyword matching, and rule-based matching to evaluate the matching degree of the job seeker with each job. For employer demand analysis and job seeker matching, for the recruitment information provided by the employer, the large model deeply analyzes its requirements for the job seeker's education, age, and ability, searches for candidates who meet the requirements in the job seeker database, and also uses the multi-dimensional matching algorithm to evaluate the matching degree of the job seeker with the recruitment position.

[0037] In one embodiment, the database management module includes a job database, a job seeker database, and an enterprise database. The job database, job seeker database, and enterprise database are respectively used to store job information, personal information, and enterprise information.

[0038] In one embodiment, the result display and feedback module is responsible for receiving the matching result data transmitted by the large model analysis module, and performing careful formatting processing and visual display according to the display characteristics of different terminals and user needs.

[0039] An intelligent recruitment method based on artificial intelligence multi-terminal application includes the following steps:

[0040] Step 1: Job seeker recruitment or employer recruitment

[0041] Step 2: Intelligent interview.

[0042] In one embodiment, in step one, the job seeker opens the application at either end, clicks the "Job Seeker Registration" button, enters the mobile phone number or email address, obtains the verification code and fills it in to complete the registration. Then, according to the system prompts, the job seeker fills in personal basic information, educational background, work experience, and job hunting intentions in sequence. During the filling process, the intelligent chat window can provide help and guidance at any time to answer the job seeker's questions about information filling. After the information filling is completed, the information interaction and processing module transmits the job seeker's information to the large model analysis module. The large model analysis module conducts a matching search in the job database based on the comprehensive situation of the job seeker, recommends suitable jobs for the job seeker, and returns the recommended job information to the job recommendation display area on the job seeker side, arranged in descending order of matching degree. The job seeker browses the list of recommended jobs, clicks on the job of interest to view the details. If the job seeker decides to apply for the job, clicks the "Apply for Job" button, and the system automatically generates a job application and sends it to the corresponding business owner. The job seeker can also further communicate with the chatbot in the intelligent chat window about job-related issues. The business owner clicks the "Business Owner Registration" button at the corresponding end, fills in the basic information of the enterprise. The business owner clicks the "Post Recruitment Job" button and collaborates with the chatbot through the intelligent chat window to fill in the recruitment job information. The chatbot will ask questions and confirm based on the partial information provided by the business owner. After the recruitment job information is posted, the large model analysis module searches for job seekers who meet the requirements in the job seeker database, and the large model analysis module recommends the matched job seekers to the business owner. The business owner views the recommended job seeker information in the job seeker recommendation display area, and clicks on the job seeker's name to view the detailed resume. The business owner selects a suitable job seeker according to the matching degree of the job seeker and his own needs, and clicks the "Contact Job Seeker" button to arrange the subsequent interview session. In step two, when both the business owner and the job seeker reach a preliminary intention, the business owner can initiate an intelligent interview in the system, select the interview time and notify the job seeker. The system generates an interview interface on the application sides of both parties. Before the interview starts, the large model generates a personalized interview question bank according to the job requirements and the situation of the job seeker, and loads it into the interview question display area. When the interview starts, the chatbot introduces the interview process and rules in the video call area in voice or text form. After the job seeker answers the questions, the chatbot uses speech recognition technology to convert the voice answer into text, and then uses natural language processing technology to analyze the answer content, evaluate the job seeker's professional knowledge level, language expression ability, and adaptability, and compare it with the job requirements to generate a real-time interview evaluation report, which is displayed on the interview interface on the business owner side, including scores of various abilities, matching degree evaluation with the job, highlights and deficiencies of the answers, etc. The business owner can supplement questions or conduct further communication in the text communication area according to the feedback of the chatbot and his own judgment. After the interview ends, the business owner records the interview result in the system, and the system automatically feeds back the interview result to the job seeker, and updates the relevant information in the job seeker database and the job database according to the interview result and process data.

[0043] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various changes or substitutions thereof, and these should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. An intelligent recruitment system based on LLM (Large Language Model) multi-terminal application, characterized by: It includes multi-terminal interactive interface module, information interaction and processing module, large model analysis module, database management module and result display and feedback module; The multi-terminal interactive interface module provides a unified user access point on the App side, WeChat applet side, Douyin applet side and PC side; The information interaction and processing module monitors various input information of multiple users in real time, encrypts the collected information and transmits it to the large model analysis module, and feeds back the processing results of the large model to the corresponding interface display module, temporarily stores commonly used and hot data, and reduces the resource consumption of repeated data acquisition and transmission; The large model analysis module, model selection and adaptation link selects the LLM large model suitable for the recruitment scenario, mainly using the Transformer model, through the core components such as self-attention mechanism, multi-head attention mechanism, feedforward neural network and position encoding, to achieve excellent performance in various generation tasks. And based on the professional knowledge, industry characteristics and common expression habits in the recruitment field, in-depth optimization training is carried out. The job seeker analysis and job matching function uses the powerful semantic understanding and analysis capabilities of the large model, as well as the powerful feature understanding capabilities of the neural network, to deeply mine and extract features of the job seeker's personal information, educational background, work experience and skills, and then conduct multi-dimensional and refined semantic and knowledge-level matching with the massive job requirements in the job database. The business owner demand analysis and job seeker matching function is based on the recruitment position information provided by the business owner. The large model accurately interprets the job seeker's education, age, professional skills and work experience, and searches and locates potential suitable candidates in the job seeker database; The database management module regularly collects job information from multiple authoritative channels and stores data on job seekers and companies; The result display and feedback module is responsible for receiving the matching result data from the large model analysis module, and formatting and visualizing it according to the display characteristics of different terminals and user needs. For the job seeker terminal, the recommended job positions are presented in a list, and for the business owner terminal, the recommended job seeker information is displayed in an orderly manner. A complete user feedback collection mechanism is established, and after the job seeker or business owner views the recommended results, specific modification opinions and suggestions are put forward. The system provides a dialogue robot for interaction, which takes the form of a chat window, and the interactive content supports multiple modes such as text, voice, picture, and video.

2. According to claim 1, the intelligent recruitment system based on LLM multi-terminal application is characterized in that: The multi-terminal interactive interface module includes a unified design principle, a job seeker interface and an enterprise owner interface. The unified design principle adopts a unified visual style and interactive design specifications on the App side, WeChat mini-program side, Douyin mini-program side and PC side. The job seeker interface: sets a striking "job seeker entrance", which enters the job seeker's exclusive interface after clicking. The interface mainly includes a personal information editing area, which guides job seekers to enter their name, age, gender, contact information, graduation school, major, educational level, work experience, certificates obtained, self-evaluation, and job search intention information in the form of a form. At the same time, an intelligent chat window is provided to interact with job seekers in an anthropomorphic image, collect information and answer questions. In addition, a job recommendation display area is provided, which presents recommended jobs in the form of a list. Each job entry displays the job title, company name, salary range, and work location information, and provides a "View Details" button, which can be clicked to expand the detailed job description, job requirements, and other information. The interface of the business owner can be entered through the "business owner entrance", including the business information registration and editing area, requiring the business owner to fill in the business name, business type, business size, business profile, contact name, contact number, and email information. There is a recruitment position release area, where business owners can post recruitment positions and use the smart chat window to assist in filling in the job title, job responsibilities, educational requirements, professional requirements, work experience requirements, skill requirements, salary and benefits, number of recruits, work location, and work schedule information. It is equipped with an intelligent chat window for communicating with business owners about recruitment needs and strategies, as well as a job seeker recommendation display area, which displays the recommended job seeker's name, photo, education, major, work experience summary, matching score and other information. Click to view the complete resume, and there is a "Contact Job Seeker" button, business owners can communicate further with one click.

3. According to claim 1, the intelligent recruitment system based on LLM multi-terminal application is characterized in that: The information interaction and processing module includes an information collection submodule, an information transmission submodule and an information caching submodule. The information collection submodule is responsible for monitoring the user input information in the intelligent chat windows and information editing areas at each end, and collecting the job seekers' personal information, job search intentions and the business owner's corporate information and recruitment requirements data in real time. The information transmission submodule encrypts the information that has been preliminarily processed and transmits it to the large model analysis module, using a safe and reliable network transmission protocol. At the same time, it receives the matching results and analysis data returned by the large model analysis module and transmits them to the corresponding interface display module. The information caching submodule establishes a local caching mechanism to cache commonly used data.

4. According to claim 1, the intelligent recruitment system based on LLM multi-terminal application is characterized in that: The large model analysis module includes model selection and adaptation, job seeker analysis and job matching, and business owner demand analysis and job seeker matching. The model selection and adaptation uses the LLM large model suitable for natural language processing tasks, and conducts targeted fine-tuning training on the large model according to the professional terms, industry knowledge, and common expression habits in the recruitment field, and builds a special recruitment field corpus, including a large number of real job descriptions, job seeker resumes, and recruitment dialogue record data for model training and optimization. The job seeker analysis and job matching receive job seeker information from the information interaction and processing module, use the large model to conduct a comprehensive analysis of it, extract the key skills, knowledge level, professional quality and other characteristics of the job seeker, and perform multi-dimensional matching with the job requirements in the job library, and use a combination of semantic similarity calculation, keyword matching, rule-based matching and other algorithms to evaluate the matching degree between the job seeker and each position. The business owner demand analysis and job seeker matching are based on the recruitment information provided by the business owner. The large model deeply analyzes the job seeker's education, age, and ability, searches for candidates who meet the requirements in the job seeker database, and also uses a multi-dimensional matching algorithm to evaluate the matching degree between the job seeker and the recruitment position.

5. According to claim 1, the intelligent recruitment system based on LLM multi-terminal application is characterized in that: The database management module includes a job database, a job seeker database and an enterprise database, and the job database, job seeker database and enterprise database are used to store job information, personal information and enterprise information respectively.

6. The intelligent recruitment system based on LLM multi-terminal application according to claim 1 is characterized by: The result display and feedback module is responsible for receiving the matching result data from the large model analysis module, and performing careful formatting and visual display according to the display characteristics of different terminals and user needs.

7. An intelligent recruitment method based on artificial intelligence multi-terminal application, supporting an intelligent recruitment system based on LLM multi-terminal application as claimed in any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: Recruitment of job seekers or business owners Step 2: Smart interview.

8. The intelligent recruitment method based on artificial intelligence multi-terminal application according to claim 7 is characterized in that: In the step 1, the job seeker opens the application on either end, clicks the "Job Seeker Registration" button, enters the mobile phone number or email address, obtains the verification code and fills it in to complete the registration, and fills in the personal basic information, educational background, work experience, and job search intention in sequence according to the system prompts. During the filling process, the intelligent chat window can provide help and guidance at any time to answer the job seeker's questions about information filling. After the information is filled in, the information interaction and processing module transmits the job seeker information to the large model analysis module. The large model analysis module performs a matching search in the job database based on the comprehensive situation of the job seeker, recommends suitable positions for the job seeker, and returns the recommended position information to the job seeker's end. The recommended position display area is arranged from high to low according to the matching degree, and the job seeker browses the recommended positions. Recommended positions list, click on the position you are interested in to view the details. If you are sure to apply for the position, click the "Apply for Position" button. The system will automatically generate a job application and send it to the corresponding business owner. Job seekers can also communicate with the chat robot on job-related issues in the intelligent chat window. Business owners click the "Business Owner Registration" button on the corresponding end to fill in the basic information of the company. Business owners click the "Publish Recruitment Position" button and fill in the recruitment position information in collaboration with the chat robot through the intelligent chat window. The chat robot will ask and confirm based on some of the information provided by the business owner. After the recruitment position information is released, the large model analysis module searches for job seekers who meet the requirements in the job seeker database. The large model analysis module recommends matching job seekers to business owners. Business owners can The recommended job seeker display area can be used to view the recommended job seeker information. Click the job seeker's name to view the detailed resume. The business owner selects the appropriate job seeker based on the job seeker's matching degree and his / her own needs and clicks the "Contact Job Seeker" button to arrange the subsequent interview. In step 2, when the business owner and the job seeker reach a preliminary intention, the business owner can initiate an intelligent interview in the system, select the interview time and notify the job seeker. The system generates an interview interface on the application side of both parties. Before the interview begins, the big model generates a personalized interview question library based on the job requirements and the job seeker's situation, and loads it into the interview question display area. At the beginning of the interview, the chatbot introduces the interview process and rules in voice or text in the video call area. After the job seeker answers the question, the chatbot Use speech recognition technology to convert voice answers into text, and then use natural language processing technology to analyze the answer content, evaluate the job seeker's professional knowledge level, language expression ability, and adaptability, and compare them with job requirements to generate a real-time interview evaluation report and display it on the interview interface of the business owner. It includes scores for various abilities, evaluation of the degree of matching with the job, highlights and shortcomings of the answer, and other information. Business owners can ask additional questions or conduct further communication in the text communication area based on the feedback from the chatbot and their own judgment. After the interview, the business owner records the interview results in the system, and the system automatically feeds back the interview results to the job seeker, and updates the relevant information in the job seeker database and job database based on the interview results and process data.

Images