Student post recommendation system and method based on intelligent model

Abstract

The invention relates to a student post recommendation system and method based on an intelligent model. The system comprises a student management module which is used for collecting and storing student resume features and job hunting intentions; wherein the student resume features comprise at least one of personal basic features, skill features, academic features and practice features; the company management module is used for collecting and storing post characteristics; wherein the post features comprise post skill features and post experience features; and the agent module comprises a pre-trained intelligent model, and is used for finding a target post matched with the resume feature and the job application intention for the expiration date by utilizing the intelligent model, or finding the expiration date corresponding to the resume feature matched with the target post for a company.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a student job recommendation system and method based on an intelligent model. Background Art

[0002] In the current job market, fresh college graduates face unprecedented challenges.

[0003] To address this situation, we propose a recruitment agent system designed specifically for recent college graduates. This system aims to help them overcome employment difficulties and improve their job search success rate through accurate job recommendations and personalized simulated interview functions.

[0004] For recent college graduates, the job search process often faces challenges such as lack of work experience, a wide range of job options, and a lack of understanding of market demand. Traditional job recommendation systems fail to fully account for these characteristics, resulting in recommended positions that may not necessarily match a graduate's actual abilities and interests. By introducing an agent based on a large model, this solution can provide personalized job recommendations based on the resume and requirements of recent graduates. Simulated interviews also help them prepare for interviews, enhancing their job search competitiveness. Summary of the Invention

[0005] The purpose of this application is to provide a student job recommendation system and method based on an intelligent model.

[0006] According to the first aspect of the embodiment of the present application, a student job recommendation system based on an intelligent model is provided, including: a student management module, used to collect and store student resume characteristics and job-seeking intentions of recent graduates; wherein the student resume characteristics include: at least one of: personal basic characteristics, skill characteristics, academic characteristics and internship characteristics; a company management module, used to collect and store job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics; an intelligent agent module, containing a pre-trained intelligent model, used to use the intelligent model to find target positions for recent graduates that match their resume characteristics and job-seeking intentions, or to find recent graduates for companies whose resume characteristics match the target positions.

[0007] Optionally, the intelligent agent module is also used to: obtain student resume files and job-seeking intentions of students; use a pre-trained intelligent model to extract student resume features from student resume files; use a pre-trained intelligent model to extract intention features of fresh graduates from job-seeking intentions.

[0008] Optionally, the intelligent agent module is also used to: respond to job search requests from recent graduates, use an intelligent model to calculate a first matching degree between student resume features and job features; if the first matching degree is greater than a first threshold, determine at least one target position found for the recent graduate.

[0009] Optionally, the intelligent agent module is also used to: respond to job search requests from recent graduates, calculate a first matching degree between student resume features and job features using an intelligent model; if the first matching degree is less than a first threshold, construct a secondary matching feature using internship features, skill features, and job search intentions; calculate a second matching degree between the secondary matching feature and job features using an intelligent model; if the second matching degree is greater than the first threshold, determine at least one target position found for the recent graduate.

[0010] Optionally, the intelligent agent module is also used to: respond to the company's recruitment request, use the intelligent model to calculate the third matching degree between the designated position and the student resume characteristics; if the third matching degree is less than the third threshold, it is determined that at least one recent graduate is found for the company.

[0011] Optionally, the system also includes an interview module for generating simulated interview questions based on the student resume characteristics and the target position characteristics; determining excellent reference samples based on the target position characteristics; and determining user capacity improvement strategies based on the simulated interview results after the students complete the virtual interview questions and the excellent reference samples.

[0012] Optionally, the interview module is used to calculate a first similarity between the personal basic characteristics and skill characteristics and the job skill characteristics; calculate a second similarity between the academic characteristics and the internship characteristics and the job experience characteristics; calculate the matching degree between the student and the target job based on the first similarity and the second similarity; and generate simulated interview questions based on the matching degree calculation results.

[0013] Optionally, the interview module is used to generate technical interview questions based on the job skill characteristics and the job experience characteristics when the matching calculation result is greater than the matching threshold; and / or generate interview questions based on the skill characteristics, the academic characteristics, and the internship characteristics.

[0014] Optionally, the interview module is used to infer the sample basic characteristics and sample skill characteristics based on the job skill characteristics; to infer the sample academic characteristics and sample internship characteristics based on the job experience characteristics; and to construct the excellent reference sample using the sample basic characteristics, the sample skill characteristics, the sample academic characteristics and the sample internship characteristics.

[0015] Optionally, the interview module is used to collect the time length, completion rate, and answer ideas of the students to complete the virtual interview questions; generate scores according to preset weight values ​​based on the time length, completion rate, and answer ideas to obtain simulated interview results.

[0016] Optionally, the interview module is used to generate skill enhancement prompts corresponding to the simulated interview questions when the answering ideas are correct, the time length exceeds the time threshold, and the completion rate is less than the completion threshold; when the answering ideas are wrong, generate learning prompts for the skill areas to which the simulated interview questions belong; use the student resume characteristics and the excellent reference samples for differentiated comparison to generate a shortcoming supplement strategy; generate the user improvement strategy based on the skill enhancement prompts, the skill area learning prompts and the shortcoming supplement strategy.

[0017] Optionally, the method further includes: inputting the student resume characteristics, the user promotion strategy, and the job trend into a job recommendation model to recommend at least one candidate job to the student to replace the target job.

[0018] According to the second aspect of the embodiment of the present application, a student job recommendation method based on an intelligent model is provided, the method comprising: collecting and storing student resume characteristics and job-seeking intentions of recent graduates; wherein the student resume characteristics include: at least one of: basic personal characteristics, skill characteristics, academic characteristics and internship characteristics; collecting and storing job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics; using a pre-trained intelligent model to find a target job for the recent graduate that matches the resume characteristics and the job-seeking intention, or to find a recent graduate for the company whose resume characteristics match the target job.

[0019] Compared to the prior art, the present application has the following advantages: a student management module for collecting and storing resume features and job-seeking intentions of recent graduates, wherein the student resume features include at least one of basic personal characteristics, skill features, academic features, and internship features; a company management module for collecting and storing job features, wherein the job features include job skill features and job experience features; and an agent module containing a pre-trained intelligent model for using the intelligent model to find target positions for recent graduates that match the resume features and job-seeking intentions, or to find recent graduates with resume features that match the target positions for the company. Through the above scheme, the intelligent model is pre-trained. Since students have no work experience, feature extraction and subsequent feature comparison focus on extracting non-work experience features. Furthermore, when matching with target positions, non-work experience features are also used, allowing recent graduates to have a greater chance of matching. This can also help recent graduates achieve a higher interview success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 A schematic diagram of the structure of the student job recommendation system based on the intelligent model provided in an embodiment of the present application.

[0021] Figure 2 A flowchart of a student job recommendation method based on an intelligent model proposed in an embodiment of the present application.

[0022] Figure 3 It is a block diagram of an electronic device according to an exemplary embodiment. DETAILED DESCRIPTION

[0023] Unless otherwise defined, the technical or scientific terms used in this specification and claims shall have the ordinary meanings understood by persons having ordinary skills in the technical field to which the invention belongs. Specific embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be noted that in the specific description of these embodiments, for the sake of clarity and clarity, this specification cannot provide a detailed description of all features of the actual embodiments. Without departing from the spirit and scope of the present invention, those skilled in the art may modify and replace the embodiments of the present invention, and the resulting embodiments are also within the scope of protection of the present invention.

[0024] like Figure 1 This is a schematic diagram of the structure of a student job recommendation system based on an intelligent model provided by the present disclosure. Figure 1 As can be seen in the figure, the system includes: a student management module 11, which is used to collect and store the resume characteristics and job-seeking intentions of recent graduates; wherein the student resume characteristics include: at least one of personal basic characteristics, skill characteristics, academic characteristics and internship characteristics.

[0025] The company management module 12 is used to collect and store job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics.

[0026] The intelligent agent module 13 includes a pre-trained intelligent model, which is used to use the intelligent model to find target positions for fresh graduates that match their resume characteristics and job-seeking intentions, or to find fresh graduates with resume characteristics that match target positions for companies.

[0027] The intelligent model mentioned here can be trained based on a large model. Fine-tuning the large model based on actual usage requirements allows the intelligent agent to be more competent in job recommendation tasks, such as recommending target positions that meet the needs of recent graduates or recommending recent graduates that meet the company's needs.

[0028] In order to efficiently match recent graduates with company openings during the recruitment process, key information must first be captured: the student's resume and the position they are interested in. This information is then fed into a pre-trained interview intelligence model to extract key features.

[0029] A student resume is typically a comprehensive document that includes a student's personal information, educational background, professional skills, project experience, internships, and more. These resume files, which may be in Word, PDF, or plain text formats, are comprehensive and designed to fully showcase a student's abilities and experience.

[0030] Students can find information about target positions through online recruitment platforms, company websites, or campus job fairs. This information typically includes the job title, responsibilities, job requirements (such as required skills and experience), location, and salary. Understanding this information helps more accurately identify the characteristics of the target position.

[0031] The job recommendation model is an intelligent tool built on big data and machine learning technology. It can automatically extract key information from resume text. For students' resume files, the model will extract the following features:

[0032] Basic personal characteristics: including the student's name, gender, age, contact information, place of origin, photo and other basic information. This information helps to establish a preliminary impression of the student.

[0033] Skill profile: This covers students' professional skills, language proficiency, soft skills (such as communication and teamwork), and hard skills. These skills are important criteria for assessing whether students are suitable for specific positions.

[0034] Academic characteristics: These include the student's educational background (e.g., academic qualifications, university of graduation, major), academic achievements (e.g., published papers, research projects participated in), GPA, etc. These characteristics reflect the student's academic foundation and research ability.

[0035] Internship Characteristics: This records the student's internship experience, including the internship company, position, responsibilities, internship duration, and evaluations received. Internship experience is a key indicator for assessing students' practical ability and professional qualities.

[0036] Extracting job characteristics from target job information: The following characteristics will be extracted from the target job information of the student's interest: Job Skill Characteristics: This identifies the professional skills, soft skills, and any specific technology or tool requirements required for the job. These skill characteristics form the basis for assessing whether the student possesses the necessary capabilities for the job.

[0037] Job experience characteristics: This includes the job's work experience requirements, such as experience in related industries, experience in specific positions, project management experience, etc. These experience requirements reflect the nature and complexity of the job and help us more accurately match students with jobs.

[0038] In addition, feature analysis and extraction can be performed based on the job-seeking intentions of fresh graduates to find the target positions that fresh graduates expect.

[0039] It's important to note that job-seeking intentions don't necessarily align perfectly with a graduate's major (the academic characteristics mentioned above). It's common for graduates to choose a specific major after the college entrance exam, but upon graduation, they often find themselves working in unrelated positions due to hobbies or circumstances.

[0040] Traditional recruitment software often recommends suitable positions based on a job seeker's work experience. This disadvantage applies to recent graduates, who often lack work experience. When submitting resumes through the same recruitment system, experienced users have a higher success rate than recent graduates. However, existing recruitment systems are even less user-friendly for recent graduates applying from different fields. In other words, recent graduates with different fields have even lower success rates.

[0041] Therefore, the disclosed technical solution not only collects the resume characteristics of recent graduates, but also their job-seeking intentions. By combining these resume characteristics and job-seeking intentions, the proposed solution recommends target positions that meet the needs of recent graduates. This disclosed solution enables more accurate and efficient job recommendations for recent graduates.

[0042] In one or more embodiments of the present disclosure, the intelligent agent module is also used to: obtain student resume files and job-seeking intentions of students; use a pre-trained intelligent model to extract student resume features from student resume files; and use a pre-trained intelligent model to extract intention features of recent graduates from job-seeking intentions.

[0043] This disclosed technical solution uses intelligent means to accurately match recent graduates with target positions. First, the system collects student resumes and job application information. The resumes are then subjected to multi-dimensional feature extraction using a pre-trained deep learning model. These features include: basic personal characteristics (identity information such as name, gender, age, and contact information), skill characteristics (hard skills such as programming languages ​​and design tools, and soft skills such as communication and collaboration), academic characteristics (institutional background, professional field, scientific research achievements, GPA, and other academic ability indicators), and internship characteristics (practical experience such as internship unit, job responsibilities, and project achievements). At the same time, for job application intentions submitted by students, the system uses a semantic analysis model to extract job skill characteristics (such as explicit skills such as specific technology stack requirements and qualification certificates) and job experience characteristics (implicit requirements such as industry background and project management experience), paying special attention to potential deviations between job application intentions and professional backgrounds—for example, students may choose to work in different fields due to changes in career interests or industry trends. Compared to traditional recruitment systems that rely heavily on work experience for matching, this solution innovatively builds a dual-feature analysis system, assessing candidates' overall capabilities through resume characteristics while capturing their career preferences through their job-seeking intentions. This effectively addresses the recommendation bias caused by recent graduates' lack of work experience. For cross-disciplinary job search scenarios, the system leverages transfer learning models to uncover cross-disciplinary correlations between skill profiles. For example, it intelligently matches the data analysis capabilities of students with science and engineering backgrounds with the quantitative requirements of financial industry positions, thereby overcoming professional barriers and achieving precise recommendations.

[0044] In one or more embodiments of the present disclosure, the intelligent agent module is also used to: respond to job search requests from recent graduates, use an intelligent model to calculate a first matching degree between student resume features and job features; if the first matching degree is greater than a first threshold, determine at least one target position found for the recent graduate.

[0045] The intelligent agent module achieves accurate job recommendations through a hierarchical matching mechanism. When a fresh graduate initiates a job request, the system first aligns the student's resume characteristics with the target job characteristics based on a pre-trained intelligent model to calculate the first matching degree. This matching degree is achieved through a multi-dimensional weighted evaluation: for example, the professional courses in the academic characteristics are associated with the technical requirements in the job skill characteristics in the knowledge graph, and the similarity is calculated by combining the project results in the internship characteristics with the project management requirements in the job experience characteristics. At the same time, the attention mechanism is introduced to dynamically adjust the weights of each dimension. If the first matching degree exceeds the preset threshold (such as 75%), the system directly outputs a list of successfully matched positions, and an accompanying visual matching report, marking key matching items such as "Python skills match job requirements 92%" or "Scientific research project experience covers machine learning scenarios required for the job."

[0046] Based on the above scheme, when recommending and matching jobs for fresh graduates, feature extraction can be performed based on other information specific to fresh graduates, such as their Chinese and African work experience (for example, basic personal characteristics, skill characteristics, academic characteristics, etc.), thereby effectively improving the probability of fresh graduates' resumes being matched.

[0047] In one or more embodiments of the present disclosure, the intelligent agent module is also used to: respond to job search requests from recent graduates, calculate a first matching degree between student resume features and job features using an intelligent model; if the first matching degree is less than a first threshold, construct a secondary matching feature using internship features, skill features, and job search intentions; calculate a second matching degree between the secondary matching feature and job features using an intelligent model; if the second matching degree is greater than the first threshold, determine at least one target position found for the recent graduate.

[0048] In practice, some recent graduates choose to apply for jobs in a different field. This means their job search intentions may differ from their major. For example, a recent graduate majoring in electrical automation may apply for a human resources position.

[0049] For recent graduates who submit resumes in a different field, not only do they lack relevant work experience, but their field of study also doesn't align with the position they're applying for. In this case, the probability of a match between their resume and the target position is very low. Therefore, to provide more accurate results for recent graduates, when matching based on conventional features, if the first match result is less than the first threshold, we use job search intent to re-match the graduate to a corresponding position.

[0050] Specifically, for scenarios where the initial matching fails to meet the criteria (that is, the first matching result is less than the first threshold), the system initiates an enhanced matching process. At this point, the intelligent model will construct a secondary matching feature matrix: on the one hand, it will deeply mine the internship characteristics, such as parsing the implicit skills in the internship description through natural language processing (such as "coordinating cross-departmental meetings" mapped to communication skills); on the other hand, it will associate skill characteristics with job-seeking intentions and model them to identify students' potential career tendency labels (such as "strong interest in the field of artificial intelligence but lack direct experience"). The secondary matching adopts a transfer learning framework to establish a cross-domain mapping between students' explicit characteristics (such as winning mathematical modeling competitions) and implicit job requirements (such as the logical thinking requirements of data analysis positions).

[0051] Based on the above public scheme, we can see that through the secondary matching method, fresh graduates can be matched with target positions that meet their needs based on their actual skills and abilities, as well as their actual job-seeking intentions. This can meet the diverse job-seeking needs of fresh graduates without work experience.

[0052] In one or more embodiments of the present disclosure, the intelligent agent module is also used to: respond to the company's recruitment request, use the intelligent model to calculate the third matching degree between the specified position and the student resume characteristics; if the third matching degree is less than the third threshold, it is determined that at least one recent graduate has been found for the company.

[0053] The intelligent agent module implements a two-way matching mechanism for enterprise recruitment needs. When an enterprise submits a job request, the system first analyzes the explicit requirements (such as professional background and skills certifications) and implicit requirements (such as team culture fit and development potential) in the job description, constructing a job profile map using a graph neural network. The intelligent model then scans the resumes of all recent graduates in the talent pool to calculate a third-order match. This process not only compares the direct match rate of skill tags but also uses latent semantic analysis to uncover implicit associations—for example, mapping the job requirement of "stress tolerance" with multi-threaded task handling experience in student club management. If the third-order match exceeds a preset threshold (e.g., 80%), the system prioritizes candidates with highly compatible characteristics and generates a talent competitiveness analysis report, noting key matching dimensions such as "87% overlap between computer science background and job skill stack" or "91% match between internship project complexity and job level requirements."

[0054] For positions that do not meet the matching requirements, the system initiates an intelligent completion strategy: based on the core competency requirements of the position, it dynamically adjusts the feature weights and expands the search radius. For example, for positions that urgently need "embedded development" experience but where talent is scarce, the system will use the transfer learning model to identify recent graduates with relevant foundations (such as electronic engineering) and the ability to learn quickly (such as online course completion records). At the same time, it introduces a corporate preference learning module, which optimizes the recommendation strategy based on historical recruitment data - if a company has hired a candidate from a different discipline in the past but with outstanding project results, the system will proactively relax professional restrictions. This two-way matching mechanism has increased the efficiency of recruiting for unpopular positions by more than 35%, especially in emerging fields such as smart manufacturing and new energy, successfully unearthing potential candidates that have been overlooked under traditional screening models.

[0055] In addition, the system also has an interview module 14, which is used to generate simulated interview questions based on the characteristics of the student's resume and the characteristics of the target position; determine excellent reference samples based on the characteristics of the target position; and determine user capacity improvement strategies based on the simulated interview results and excellent reference samples after the students complete the simulated interview questions.

[0056] The interview module 14 is also used to calculate a first similarity between personal basic characteristics and skill characteristics and job skill characteristics; calculate a second similarity between academic characteristics and internship characteristics and job experience characteristics; calculate the matching degree between the fresh graduate and the target job based on the first similarity and the second similarity; and generate simulated interview questions based on the matching degree calculation results.

[0057] The interview module 14 is further configured to generate technical interview questions based on job skill characteristics and job experience characteristics when the matching degree calculation result is greater than the matching degree threshold; and / or generate interview questions based on skill characteristics, academic characteristics, and internship characteristics.

[0058] The interview module 14 is also used to infer the basic characteristics and skill characteristics of the sample based on the job skill characteristics; to infer the academic characteristics and internship characteristics of the sample based on the job experience characteristics; and to construct an excellent reference sample using the basic characteristics, skill characteristics, academic characteristics and internship characteristics of the sample.

[0059] The specific implementation of the interview module 14 will be described in detail through the following examples. Figure 2 This is a flow chart of a method for recommending student positions based on an intelligent model provided by the present disclosure. This method can be applied to a server (including a local server or a cloud server) to Figure 2 The method includes the following steps:

[0060] Step 201: Collect and store student resume characteristics and job-seeking intentions of recent graduates; wherein the student resume characteristics include: at least one of: basic personal characteristics, skill characteristics, academic characteristics and internship characteristics.

[0061] Step 202: Collect and store job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics.

[0062] Step 203: Use the pre-trained intelligent model to find a target position for the fresh graduate that matches the resume features and the job-seeking intention, or find a fresh graduate whose resume features match the target position for the company.

[0063] It should be noted that during the matching process, if the student's resume features extracted from the resume are not matched to the appropriate target position, it may be because the features of the resume are not obvious or the wording is not standardized. In this case, we can select similar excellent reference samples to help the student modify and polish the resume. Of course, we can also use large models for polishing. This makes the modified resume features more obvious, which facilitates the use of intelligent models to quickly and accurately match the target position features.

[0064] Step 204: Utilize the pre-trained intelligent model to obtain student resume features and target position features.

[0065] Step 205: Generate simulated interview questions based on the student resume characteristics and the target position characteristics.

[0066] Step 206: Determine excellent reference samples based on the target position characteristics.

[0067] Step 207: Determine a user capability improvement strategy based on the simulated interview results after the students complete the virtual interview questions and the excellent reference samples.

[0068] The pre-trained interview intelligent model mentioned here is trained based on the Word2Vec model. Its training process covers several key steps to ensure that the model can efficiently extract student resume features and target position features. First, the input text data is preprocessed, including word segmentation (for Chinese text, tools such as jieba can be used) and stop words are removed to build a clear vocabulary. Subsequently, a suitable model architecture is selected, such as the Skip-Gram model (predicting the target word through the context) or the CBOW model (predicting the context through the target word), and key hyperparameters are set, such as the word vector dimension (usually 100 to 300 dimensions are selected according to the application scenario), the context window size (generally 5 to 10), and the minimum word frequency (to filter out overly rare words). During the training process, stochastic gradient descent (SGD) or other optimization algorithms (such as Adam) are used for iterative optimization until the model converges, and finally the trained word vectors and model parameters are saved for subsequent use.

[0069] In terms of dataset selection, model training utilizes a diverse range of text resources. Core datasets include university student resumes, which contain rich information on students' educational backgrounds, internship experiences, and skills, enabling the model to accurately identify student competencies. The target job description dataset covers job requirements and responsibilities across various industries, helping the model understand the core demands of each position. Furthermore, to expand the vocabulary and enhance the model's understanding of common language, a general text dataset was introduced. Its large-scale corpus enhances the model's semantic expression capabilities, making it more widely applicable in simulated interview scenarios.

[0070] In order to further optimize the performance of the model, the present invention adopts a variety of strategies. During the training process, negative sampling technology is introduced to reduce the amount of calculation by randomly selecting negative samples, while improving the training efficiency and the convergence speed of the model. Hyperparameters such as word vector dimension, context window size, etc. are adjusted through experiments to achieve optimal performance. After the training is completed, the word vectors are normalized to enhance the generalization ability of the model so that it can better adapt to different types of text input. In addition, considering the limitations of a single model, the present invention also explores model fusion technology, combining the Word2Vec model with a more advanced language model (such as BERT, GPT), and using its pre-trained deep semantic features to further enhance the model's ability to understand complex semantics, thereby more accurately generating simulated interview questions and formulating user capability improvement strategies.

[0071] Furthermore, since students lack work experience, the model can focus on extracting and focusing on skills and internship-related features to compensate for this lack of work experience. Furthermore, it can also provide students with a mock interview function, allowing them to simulate real interview content in advance, improve their interview experience, and thus help them better pass the job interview.

[0072] To further enhance the system's intelligence, this system incorporates agent orchestration and Retrieval-Augmented Generation (RAG) technology. Through agent orchestration, the system can break down complex tasks into multiple subtasks and assign them to different agents for collaborative completion. For example, a resume parsing agent extracts key information from student resumes, while a question generation agent generates simulated interview questions based on the extracted features. These agents' collaborative work improves the system's efficiency and flexibility.

[0073] At the same time, RAG technology retrieves relevant information from external knowledge bases and provides it as context to the language model, thereby generating more accurate and informative interview questions and feedback. For example, when generating interview questions, RAG technology can retrieve information related to the target position, such as the latest industry trends and skill requirements, from the knowledge base to ensure the timeliness and accuracy of the question content.

[0074] Student resume features are key information points extracted from the resumes submitted by students through the interview intelligent model, including but not limited to academic qualifications, majors, internship experience, academic ability, professional skills, personal projects, awards, club activities, social activity experience, etc.

[0075] The target job characteristics mentioned here refer to the key requirement information extracted from the job description for a specific recruitment position, including job responsibilities, required skills, job experience requirements, personality trait matching, etc.

[0076] Based on the target position requirements and student resume characteristics, automatically generated interview questions are designed to assess students' professional skills, work experience compatibility, problem-solving skills, logical thinking, and other aspects. This provides students with a simulated interview opportunity. For example, the input is the student resume characteristics obtained in step 201 and the target position characteristics. Combining the key requirements of the position characteristics with the actual experience in the student's resume, interview questions are generated through rule matching or deep learning algorithms. The model outputs a set of customized simulated interview questions covering multiple dimensions, including technical and behavioral interviews.

[0077] When generating simulated interview questions, the system leverages RAG technology to retrieve knowledge base content relevant to the target position and, combined with the skills and internship experience in the student's resume, generates more targeted interview questions. For example, for technical positions, the system can retrieve the latest technology trends and skill requirements to generate relevant technical interview questions. Furthermore, agent orchestration technology ensures the efficiency and accuracy of question generation, enabling the rapid generation of high-quality interview questions through the collaborative work of multiple agents.

[0078] The excellent reference samples mentioned here refer to the resumes, interview performance records, or work achievements of individuals who have performed well in the position in the past. These can serve as a benchmark for evaluating the performance of the applicant. Alternatively, they can be virtual excellent candidates constructed using a model. In one alternative, the characteristics of the target position and the resumes of outstanding students within the company or industry are combined to select the best candidate sample that best matches the current recruitment needs.

[0079] Students’ performance scores after answering simulated interview questions in a virtual environment may include technical accuracy, answer fluency, non-verbal communication and other aspects.

[0080] Finally, based on a comparison of mock interview results with excellent reference samples, students are provided with targeted capacity-building recommendations, which may include skills training, project practice, and soft skills enhancement. By comparing students' performance with excellent reference samples, gaps are identified. A personalized capacity-building plan is then developed for students, including recommended learning resources, practical opportunities, and feedback mechanisms, aiming to help them bridge the gap between their performance and job requirements.

[0081] When generating competency improvement strategies, the system leverages RAG technology to retrieve content relevant to students' competency gaps, such as recommended learning resources and industry training courses. Furthermore, agent orchestration technology ensures efficient and personalized feedback and suggestions. For example, in a student job recommendation system based on intelligent models, the feedback generation agent analyzes the gaps between students' performance data in mock interviews (such as technical accuracy, answer fluency, and problem-solving skills) and the characteristics of excellent reference samples. This allows the agent to accurately identify students' competency gaps and provide specific improvement suggestions. For example, for students with limited technical skills, the feedback generation agent can recommend relevant learning resources, design skill practice tasks, and even provide expert guidance. For students with less fluent answers, it recommends language training tools and simulated practice scenarios, and provides detailed feedback and corrections. Furthermore, the feedback generation agent can combine RAG technology to search external knowledge bases and generate personalized improvement plans for students, including learning path planning, recommended practical opportunities, and career planning advice. This helps students systematically improve their competency, narrow the gap between them and their target positions, and better adapt to future career challenges.

[0082] Through the above steps, students' job adaptability can be evaluated efficiently and accurately, and specific growth paths can be provided, which not only improves recruitment efficiency but also promotes students' personal career development.

[0083] In one or more embodiments of the present disclosure, generating simulated interview questions based on the student resume characteristics and the target position characteristics includes: using the personal basic characteristics and skill characteristics to calculate a first similarity with the position skill characteristics; using the academic characteristics and the internship characteristics to calculate a second similarity with the position experience characteristics; calculating the matching degree between the student and the target position based on the first similarity and the second similarity; and generating simulated interview questions based on the matching degree calculation result.

[0084] During the recruitment process, to assess the student's match with the target position and generate mock interview questions accordingly, you can take the following detailed steps:

[0085] First, the basic personal characteristics and skill characteristics in the student's resume are compared with the job skill characteristics to calculate the first similarity. Basic personal characteristics, such as the student's professional background and language proficiency, can initially reflect whether they meet the basic requirements for the position. Skill characteristics, such as the student's professional skills and software operation ability, are directly related to the skill requirements of the position. By matching these characteristics with the job skill characteristics, we can obtain a preliminary assessment result based on skill similarity, which is the first similarity.

[0086] Next, the student's academic and internship characteristics are compared with their job experience characteristics to calculate a second similarity. Academic characteristics, such as a student's academic background, research focus, and academic achievements, reflect their academic foundation and research capabilities, which are particularly important in certain positions. Internship characteristics, such as internship experience, internship position, and internship achievements, directly reflect their practical experience and professional qualities. By matching these characteristics with the job experience characteristics, we can obtain an experience-based similarity assessment result, namely the second similarity.

[0087] After obtaining the first and second similarities, a weighting method is used to combine these two similarities to determine the degree of match between the student and the target position. Weighting can be determined based on factors such as the specific requirements of the position, the company's recruitment strategy, and industry practices. For example, for technology-intensive positions, skill characteristics may be given a higher weight; for positions that emphasize practical experience and professional qualities, internship characteristics may be given a higher weight.

[0088] Finally, based on the match calculation results, simulated interview questions can be generated. These questions should be closely tailored to the job's skill and experience requirements, as well as the student's resume profile, aiming to comprehensively assess the student's professional abilities, practical experience, professional qualities, and potential for advancement. The type and difficulty of the questions should be tailored to the student's specific circumstances and the specific requirements of the position to ensure a fair and effective interview.

[0089] Through the above steps, students can be accurately matched with target positions, and targeted mock interview questions can be generated accordingly to provide strong support for the subsequent interview process.

[0090] In one or more embodiments of the present disclosure, generating simulated interview questions based on the matching calculation result includes: when the matching calculation result is greater than a matching threshold, generating technical interview questions based on the job skill characteristics and the job experience characteristics; and / or generating interview questions based on the skill characteristics, the academic characteristics, and the internship characteristics.

[0091] During the recruitment process, if the calculated match between a student and the target position exceeds a preset match threshold, it indicates that the student has high potential adaptability and is suitable for the next interview stage. At this point, appropriate interview questions can be generated based on the specific requirements of the position and the characteristics of the student's resume.

[0092] First, if the match is high and the position clearly requires technical skills and practical experience, technical interview questions can be generated based on the skill and experience characteristics of the position. These questions will focus on assessing the student's mastery of professional skills, technical problem-solving ability, project-based experience, and understanding of industry knowledge. For example, for software development positions, programming, algorithm, or system design questions can be designed to test students' programming skills, algorithm understanding, and system design thinking. For marketing positions, case analysis, market research, or marketing strategy development questions can be designed to test students' market analysis, strategy formulation, and execution capabilities.

[0093] Secondly, in addition to technical interview questions, non-technical interview questions can also be generated based on students' skill profiles, academic profiles, and internship experiences. These questions aim to gain a more comprehensive understanding of students' overall qualities, personal qualities, career plans, and teamwork abilities. For example, based on students' skill profiles, scenario-based simulation questions can be designed to assess their communication, coordination, and adaptability; based on their academic profiles, academic discussion questions can be designed to assess their academic literacy and research capabilities; and based on their internship experiences, experience sharing questions can be designed to assess their practical experience, professionalism, and teamwork abilities.

[0094] When generating interview questions, it's important to pay attention to their relevance and effectiveness. Questions should be closely aligned with the job requirements and the student's resume to ensure an accurate assessment of their abilities and potential. Furthermore, the difficulty and type of questions should be tailored to the student's specific situation and the specific requirements of the position, avoiding overly simple or overly complex questions that could compromise the fairness and effectiveness of the interview.

[0095] Finally, the generated interview questions need to undergo rigorous review and testing to ensure their quality meets the requirements of the recruitment process. Experts in related fields or experienced HR personnel can be invited to review and evaluate the questions to ensure their accuracy and effectiveness. Simulated interviews can also be conducted to evaluate their performance and effectiveness in actual interviews.

[0096] Through the above steps, targeted and effective interview questions can be generated based on the matching calculation results, providing strong support for the subsequent interview process, thereby helping students and positions to achieve a better match.

[0097] In one or more embodiments of the present disclosure, determining an excellent reference sample based on the target position characteristics includes: inferring sample basic characteristics and sample skill characteristics based on the position skill characteristics; inferring sample academic characteristics and sample internship characteristics based on the position experience characteristics; and constructing the excellent reference sample using the sample basic characteristics, the sample skill characteristics, the sample academic characteristics, and the sample internship characteristics.

[0098] In the process of building an excellent reference sample, we first need to conduct an in-depth analysis based on the skills and experience characteristics of the position, so as to infer the basic characteristics, skills characteristics, academic characteristics, and internship characteristics of the sample that are highly matched with the position. The following is a detailed explanation of this process:

[0099] First, we conduct a detailed analysis of the job skill profile. This typically includes the required professional knowledge, technical skills, and software proficiency. Through a comprehensive analysis of these skill profiles, we can infer basic sample characteristics, such as the student's professional background and educational level. These foundational characteristics serve as the foundation for their mastery of the skills required for the job. Furthermore, we can further infer specific skill characteristics, such as the specific technical capabilities and project experience that students should possess, which are direct requirements for the job.

[0100] Next, we infer the sample's academic characteristics based on job experience characteristics. These typically include internships, project participation, and industry experience. By analyzing these experience characteristics, we can infer the sample's academic characteristics, such as the student's research achievements and academic paper publications. These academic characteristics reflect the student's in-depth research and academic accomplishments in the relevant field. Furthermore, we can infer the sample's internship characteristics, such as the internship position, duration, and achievements expected of the student. These internship characteristics are crucial for assessing the student's practical experience and professional qualities.

[0101] After inferring these characteristics, we begin constructing a sample of excellent candidates. This sample is based on job requirements, industry practices, and the student's actual situation, creating an idealized image of a student that is highly compatible with the position. During this construction process, we organically integrate the sample's basic characteristics, skills, academic profile, and internship experience to form a comprehensive, three-dimensional profile. This sample not only possesses the skills and experience required for the position, but also demonstrates excellent academic and professional qualities.

[0102] To ensure the accuracy and validity of excellent reference samples, they must be rigorously verified and revised. This can be achieved through communication with industry experts and experienced HR personnel to understand the actual requirements of the position and industry practices, and to make necessary adjustments and optimizations to the sample. Furthermore, historical recruitment data, industry reports, and other resources can be used to further verify and refine the sample.

[0103] Ultimately, the resulting excellent reference samples will become a crucial basis for the recruitment process. They can help more accurately assess students' abilities and potential, providing strong support for the design of interview questions and the formulation of hiring decisions. Furthermore, these excellent reference samples can serve as a basis for training and development plans, helping students better adapt to job requirements and enhance their professional qualities and overall abilities.

[0104] In one or more embodiments of the present disclosure, the method for generating the simulated interview results includes: collecting the time length, completion rate, and answer ideas of the students to complete the virtual interview questions; generating scores according to preset weight values ​​based on the time length, completion rate, and answer ideas to obtain the simulated interview results.

[0105] During the mock interview process, a series of meticulous data collection and evaluation methods were employed to generate comprehensive and objective interview results. The following is a detailed explanation of how the mock interview results were generated:

[0106] First, comprehensive data on students' performance in the virtual interview will be collected. This data primarily covers three aspects: first, the length of time it takes students to complete the virtual interview, i.e., the total time spent from the start to the end; second, the completion rate, i.e., the ratio of the number of questions completed by students to the total number of questions, which reflects the student's progress and degree of completion; and third, the student's approach to answering the questions, which requires a comprehensive assessment of the student's expression, logical structure, and problem-solving methods during the answering process, reflecting their thinking style and problem-solving strategies.

[0107] After collecting these data, they will be scored according to the preset weight values. The setting of the weight values ​​is based on an in-depth understanding of the job requirements and a precise grasp of the interview evaluation criteria. For the length of time, a certain time range may be given as a standard. Exceeding or falling below this time range will result in corresponding deductions or additions to evaluate the student's answering efficiency and time management ability. For the completion rate, a basic completion rate requirement will be set. Students who fail to meet the requirements will be deducted points, and students who exceed the requirements will receive extra points to encourage students to try their best to complete all questions. For answering ideas, points will be given based on the difficulty of the questions, the clarity of the students' logic, the innovation of the answers, etc., to evaluate students' thinking ability, problem-solving ability and innovation ability.

[0108] During the scoring process, calculations will be strictly based on pre-set weights to ensure that every piece of data is evaluated fairly and objectively. Weights may vary depending on the position. For example, for technical positions, more emphasis may be placed on answering strategies and completion rates, as these two aspects more directly reflect a student's technical and problem-solving abilities. For management positions, however, more attention may be paid to time and completion rates, as these two aspects reflect a student's time management skills and work efficiency.

[0109] Ultimately, the mock interview results are generated based on the scoring. This result is more than just a simple score; it's a comprehensive, objective assessment report, encompassing all the data from the student's virtual interview questions and their corresponding scores. This report provides a deeper understanding of the student's abilities and potential, providing strong support for subsequent hiring decisions. It also serves as a reference for students to self-assess and improve themselves, helping them better understand their strengths and weaknesses, thereby becoming more confident and prepared for future job searches.

[0110] In one or more embodiments of the present disclosure, a user capability improvement strategy is determined based on the simulated interview results after the students complete the virtual interview questions and the excellent reference samples, including: when the answering idea is correct, the time length exceeds the time threshold, and the completion rate is less than the completion threshold, a skill enhancement prompt corresponding to the simulated interview question is generated; when the answering idea is wrong, a learning prompt for the skill field to which the simulated interview question belongs is generated; a differentiated comparison is made using the student resume characteristics and the excellent reference samples to generate a shortcoming supplement strategy; and the user improvement strategy is generated based on the skill enhancement prompt, the skill field learning prompt, and the shortcoming supplement strategy.

[0111] In the subsequent processing of the mock interview, a series of targeted improvement strategies will be generated based on the student's answer performance and the preset standards. The following is a detailed explanation of this process:

[0112] First, we carefully analyze the student's approach to the question. If the student's approach is correct, it means they understand the core requirements of the question and are able to answer it logically. However, at this point, we also need to consider two other factors: time and completion rate.

[0113] If a student's time exceeds the preset time threshold, this usually means that they may have been overly cautious or encountered difficulties in understanding the questions, resulting in a slow answering speed. At the same time, if their completion rate is still less than the preset completion threshold, that is, they failed to complete all questions, then this indicates that they have deficiencies in time management and answering efficiency. In this case, skill-enhancing tips corresponding to the simulated interview questions will be generated, suggesting that students improve their answering speed and efficiency while maintaining the correct answering mindset. For example, through more practice to familiarize themselves with question types and problem-solving techniques, or through time management training to improve their sense of rhythm when answering questions.

[0114] On the other hand, if students' answering strategies are incorrect, this indicates they may have misunderstood the question or lack relevant knowledge and skills. In this case, learning tips will be generated for the skill areas to which the mock interview questions belong. These tips will clearly indicate where students are missing knowledge or understanding, and will suggest ways to address these deficiencies by studying relevant materials, attending training courses, or seeking professional guidance.

[0115] In addition to the aforementioned performance-based tips, the system also compares students' resume characteristics with those of excellent reference samples. This comparison clearly identifies areas where students differ from the excellent reference samples, representing their weaknesses. Specific strategies to address these weaknesses are generated, such as suggesting students participate in relevant projects or internships to gain practical experience, or enhance their professional skills through advanced courses or professional certifications.

[0116] Finally, the aforementioned skill enhancement tips, skill area learning tips, and strategies for addressing weaknesses are integrated into a comprehensive user improvement strategy. This strategy not only identifies specific issues students exhibited during mock interviews but also provides specific suggestions for improvement, aiming to help students comprehensively enhance their abilities and competitiveness, thereby fully preparing them for future job searches.

[0117] In one or more embodiments of the present disclosure, the method further includes: inputting the student resume characteristics, the user promotion strategy, and the job trend into a job recommendation model to recommend at least one candidate job to the student to replace the target job.

[0118] To recommend candidate positions that match students' abilities, development needs, and market trends, we use students' resume characteristics, user development strategies, and job trends as key inputs into a carefully designed job recommendation model. The following is a detailed explanation of this process:

[0119] First, we collect and organize the student's resume characteristics, including but not limited to the student's education background, professional skills, work experience, personal interests, career plans, etc. This information constitutes a comprehensive portrait of the student's personal abilities and background, which is the basis for job recommendations.

[0120] Next, we take into account the user improvement strategies generated earlier. These strategies not only reflect the student's current skill level deficiencies and potential for improvement, but also implicitly capture their aspirations and aspirations for future career development. By matching improvement strategies with job requirements, we can more accurately predict which positions students are suitable for, as well as the challenges and growth opportunities they may face in those roles.

[0121] At the same time, we closely monitor and analyze job trends, including industry development trends, changes in job requirements, and the impact of emerging technologies. Job trend data provides an intuitive reflection of market dynamics and industry prospects, helping to predict which jobs will be more attractive and which skills will be more popular in the future.

[0122] After obtaining this information, it is fed into a job recommendation model. This model, built on machine learning or deep learning algorithms, automatically analyzes and discovers correlations and patterns within the data. The model comprehensively considers the student's resume characteristics, advancement strategies, and job trends. Through a complex calculation process, it recommends a series of candidate positions that align with the student's abilities, needs, and market trends.

[0123] During the recommendation process, the model also considers job diversity, ensuring that recommended positions not only match students' personal characteristics but also meet their expectations for work environment, salary, and career development. Furthermore, the model ranks positions based on their competitive strength and the students' competence, prioritizing those positions that are more likely to be obtained and more beneficial to their personal development.

[0124] Ultimately, the job recommendation model will output a list of at least one candidate position, each with detailed job descriptions, job requirements, salary ranges, and other information, allowing students to select positions based on their interests and career plans. This recommendation process not only improves the accuracy and efficiency of job matching but also provides students with more personalized and targeted career development guidance.

[0125] 1. Resume screening and matching advantages:

[0126] Existing solutions: Rely on keyword matching or simple rule engines, which fail to understand the context of resumes and tend to overlook candidates' actual capabilities. They are unable to process unstructured data (such as free-text resumes), resulting in inaccurate screening results.

[0127] This application uses AI-driven resume feature extraction technology to extract multi-dimensional features (personal foundation, skills, academic achievements, internships, etc.). Through interview intelligent models, we deeply understand the resume context and identify potential capabilities, going beyond keyword matching. For example, even if the resume does not explicitly mention "Java development", the system can infer relevant skills through project experience, improving screening accuracy and efficiency.

[0128] 2. Advantages of interview question generation:

[0129] The existing solution used a fixed interview question bank, which had a poor match between the questions and the positions and could not be customized based on the candidate's background. For example, all candidates for the "Software Development" position received the same questions, making it impossible to distinguish between front-end and back-end skill sets.

[0130] This application dynamically generates personalized interview questions based on the candidate's resume and position characteristics. Through match calculations, technical and behavioral interview questions are generated to ensure a close match between the candidate's skills and the position requirements. For example, candidates with extensive front-end development experience will receive more front-end-related questions, improving the accuracy of the interview assessment.

[0131] 3. Interview result analysis and ability improvement strategy advantages:

[0132] Existing solutions only provide simple scores or pass / fail results, without analyzing details like answering strategies and time management. They also lack targeted skills development suggestions, making it difficult for candidates to identify their weaknesses.

[0133] This application analyzes multiple dimensions of data, including answering strategies, completion rates, and duration, to generate detailed mock interview results. Based on performance, targeted improvement strategies are generated, such as tips for strengthening skills and areas of expertise. By comparing candidates with excellent reference samples, strategies for addressing weaknesses are developed to help them comprehensively improve their abilities.

[0134] 4. Recommended advantages of the position:

[0135] Existing systems rely on simple rules or keyword matching, failing to consider candidates' career plans or skills development needs. Recommended positions don't match actual demand, resulting in low accuracy.

[0136] This application combines resume characteristics, skill development strategies, and job trends to generate a personalized job recommendation list. This list considers not only current skill levels but also future career development needs. For example, it recommends positions related to the candidate's skill development areas. It also recommends high-potential positions based on industry trends to help candidates seize market opportunities.

[0137] An embodiment of the present application also proposes an electronic device, comprising a processor and a memory; the memory is used to store a computer program executable by the processor; the processor is used to execute the computer program in the memory to implement the student job recommendation method based on the intelligent model described in any of the above embodiments.

[0138] An embodiment of the present application also proposes a computer-readable storage medium, which, when the executable computer program in the storage medium is executed by a processor, can implement the student job recommendation method based on the intelligent model described in any of the above embodiments.

[0139] Regarding the apparatus in the above embodiment, the specific manner in which the processor performs operations has been described in detail in the embodiment of the method, and will not be elaborated on here.

[0140] Figure 3 9 is a block diagram of an electronic device according to an exemplary embodiment. For example, the electronic device 900 can be provided as a server. Figure 3The device 900 includes a processing component 922, which further includes one or more processors, and a memory resource represented by a memory 932 for storing instructions executable by the processing component 922, such as an application. The application stored in the memory 932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 922 is configured to execute the instructions to perform the above-mentioned method for student job recommendation based on the intelligent model.

[0141] The device 900 may also include a power supply component 926 configured to perform power management of the device 900, a wired or wireless network interface 950 configured to connect the device 900 to a network, and an input / output (I / O) interface 958. The device 900 may operate based on an operating system stored in the memory 932, such as Windows Server™, MacOS X™, Unix™, Linux™, FreeBSD™, or the like.

[0142] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 932 including instructions, which can be executed by the processing component 922 of the device 900 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0143] In the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance. The term "plurality" refers to two or more, unless otherwise clearly defined.

[0144] The above description of the embodiments is intended to facilitate understanding and application of the present application by those skilled in the art. It will be apparent that those skilled in the art can readily make various modifications to these embodiments and apply the general principles described herein to other embodiments without expending any creative effort. Therefore, the present application is not limited to the embodiments described herein, and improvements and modifications made by those skilled in the art based on the disclosure of this application without departing from the scope and spirit of this application are within the scope of this application.

Claims

1. A student job recommendation system based on an intelligent model, characterized in that: The system comprises: A student management module is used to collect and store the resume characteristics and job-seeking intentions of recent graduates; wherein the student resume characteristics include at least one of: basic personal characteristics, skill characteristics, academic characteristics, and internship characteristics; The company management module is used to collect and store job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics; The intelligent agent module includes a pre-trained intelligent model, which is used to use the intelligent model to find a target position for the recent graduate that matches the resume characteristics and the job-seeking intention, or to find a recent graduate whose resume characteristics match the target position for the company.

2. The system according to claim 1, wherein: The intelligent agent module is further used to: Obtain student resume files and students' job search intentions; Extracting the student resume features from the student resume file using a pre-trained intelligent model; A pre-trained intelligent model is used to extract the intention features of fresh graduates from the job-seeking intentions.

3. The system according to claim 1, wherein: The intelligent agent module is further used to: In response to a job application request from a recent graduate, using the intelligent model to calculate a first matching degree between the student resume features and the job features; If the first matching degree is greater than a first threshold, it is determined that at least one target position is found for the graduate.

4. The system according to claim 1, wherein: The intelligent agent module is further used to: In response to a job application request from a recent graduate, using the intelligent model to calculate a first matching degree between the student resume features and the job features; If the first matching degree is less than a first threshold, constructing a secondary matching feature using the internship feature, the skill feature, and the job-seeking intention; Calculating a second matching degree between the secondary matching feature and the job feature using the intelligent model; If the second matching degree is greater than the first threshold, it is determined that at least one target position is found for the graduate.

5. The system according to claim 1, wherein: The intelligent agent module is further used to: In response to a company's recruitment request, using the intelligent model to calculate a third matching degree between a designated position and the student resume characteristics; If the third matching degree is less than a third threshold, it is determined that at least one fresh graduate is found for the company.

6. The system according to any one of claims 1 to 5, characterized in that The system also includes an interview module for generating simulated interview questions based on the student resume characteristics and the target position characteristics; Determine excellent reference samples based on the characteristics of the target position; Determine a user capability improvement strategy based on the mock interview results after the students complete the mock interview questions and the excellent reference samples.

7. The system according to claim 6, characterized in that The interview module is further configured to calculate a first similarity between the personal basic characteristics and the skill characteristics and the job skill characteristics; Calculating a second similarity between the academic feature and the internship feature and the job experience feature; Calculating a matching degree between the fresh graduate and the target position based on the first similarity and the second similarity; Generate simulated interview questions based on the matching degree calculation result.

8. The system according to claim 7, characterized in that The interview module is further configured to generate technical interview questions based on the job skill characteristics and the job experience characteristics when the matching degree calculation result is greater than a matching degree threshold; and / or, Interview questions are generated based on the skill characteristics, the academic characteristics, and the internship characteristics.

9. The system according to claim 6, wherein: The interview module is further used to infer sample basic characteristics and sample skill characteristics based on the job skill characteristics; Based on the job experience characteristics, the sample academic characteristics and sample internship characteristics are inferred; The excellent reference sample is constructed using the sample basic characteristics, the sample skill characteristics, the sample academic characteristics and the sample internship characteristics.

10. A student job recommendation method based on an intelligent model, characterized in that: The method comprises: Collect and store the resume characteristics and job-seeking intentions of recent graduates; wherein the resume characteristics include at least one of: basic personal characteristics, skill characteristics, academic characteristics, and internship characteristics; Collect and store job characteristics; wherein the job characteristics include: job skill characteristics and job experience characteristics; The pre-trained intelligent model is used to find a target position for the fresh graduate that matches the resume features and the job-seeking intention, or to find a fresh graduate whose resume features match the target position for the company.

Images