Artificial intelligence-driven recruitment and interview analysis system and method

Abstract

The invention provides an artificial intelligence-driven recruitment and interview analysis system and method and an artificial intelligence-driven recruitment and interview analysis system. The artificial intelligence-driven recruitment and interview analysis system comprises a resume screening module for obtaining resumes of applicants, calculating the matching degree between the resumes and preset post requirements, and screening out the applicants whose resume matching degree is greater than a preset matching degree threshold; the interview analysis module is used for sending the video interview invitation to the applicant whose resume matching degree is greater than a preset threshold value, obtaining an interview video of the applicant, and obtaining an interview score; the comprehensive analysis module is used for generating a comprehensive score according to the resume matching degree and the interview score, screening out applicants with the comprehensive score greater than a preset comprehensive score threshold value, and forming a comprehensive score queue; and the pushing module is used for sending the in-job notice to the selected applicant if the recruiter selects the applicant in the comprehensive score queue. According to the invention, through the sorting of the comprehensive score queue, a recruiter is helped to quickly identify the optimal candidate, and the scientificity of recruitment decision is improved.

Description

Technical Field

[0001] The present invention relates to human resource management technology, and in particular to an artificial intelligence-driven recruitment and interview analysis system and method. Background Art

[0002] In the traditional recruitment and interview process, companies usually rely on the experience of human resources managers to evaluate the qualifications and capabilities of candidates. Although modern technology has played a certain role in the recruitment process, such as resume screening tools and online interview platforms, the overall recruitment process is still highly dependent on human intervention. The existing automated screening system mainly focuses on the processing of resume data, using keyword matching and basic rule engines for preliminary screening. However, this method can only rely on the resume information provided by the candidate, and cannot fully evaluate the candidate's actual ability and its match with the position. Summary of the invention

[0003] In view of the deficiencies of the prior art, the present invention provides an artificial intelligence driven recruitment and interview analysis system and method.

[0004] The artificial intelligence driven recruitment and interview analysis system provided by the present invention includes the following:

[0005] Resume screening module: used to obtain applicants' resumes, parse them, calculate the matching degree between the resumes and the preset job requirements, and screen out applicants whose resume matching degree is greater than the preset matching degree threshold;

[0006] Interview analysis module: used to send video interview invitations to candidates whose resume matching degree is greater than the preset threshold. If the candidate accepts the video interview invitation, he / she can click to enter the interview start page. After the interview video is recorded, the interview video of the candidate is obtained and semantic analysis is performed on the interview video to obtain the interview score.

[0007] Comprehensive analysis module: used to generate comprehensive scores based on resume matching and interview scores, screen out candidates whose comprehensive scores are greater than the preset comprehensive score threshold, and form a comprehensive score queue, in which the candidates with the highest scores are ranked first;

[0008] Push module: used to push the comprehensive score queue to the recruiter. If the recruiter selects the applicant in the comprehensive score queue, the employment notice will be sent to the selected applicant.

[0009] Preferably, the resume screening module includes:

[0010] Resume acquisition unit: used to obtain the resume of applicants;

[0011] Resume parsing unit: used to extract key information from a resume using natural language processing, where the key information in the resume includes personal information, work experience, educational background, and skills and expertise;

[0012] Job requirement matching degree calculation unit: used to calculate the matching degree between the applicant's resume and the job requirements based on the preset job requirements and key information in the resume;

[0013] Matching degree screening unit: used to screen out applicants whose matching degree is greater than a preset matching degree threshold according to the calculated matching degree;

[0014] Data transmission unit: used to transmit the resume and matching degree of the applicant whose matching degree is greater than the preset matching degree threshold to the interview analysis module.

[0015] Preferably, the job requirement matching degree calculation unit includes

[0016] Position requirement extraction subunit: used to extract key position information required by the preset position;

[0017] Resume information acquisition subunit: used to acquire resume key information from the resume parsing unit;

[0018] Text preprocessing subunit: used to preprocess key job information and resume information;

[0019] Feature vectorization subunit: used to convert job requirements and resume content into feature vectors using vectorization methods;

[0020] Matching degree calculation subunit: used to calculate the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain the matching degree;

[0021] Matching degree value normalization processing unit: used for normalizing the calculated matching degree;

[0022] Comprehensive matching value generation unit: used to perform weighted processing on the calculated matching degree according to the preset weights of different parts of the job requirements to generate the final matching degree.

[0023] Preferably, the interview analysis module includes

[0024] Video interview invitation sending unit: used to send video interview invitations to applicants whose resume matching degree is greater than a preset threshold, and record the invitation status;

[0025] Interview video acquisition unit: used to acquire the interview video of the applicant through the built-in video recording function after the applicant accepts the video interview invitation;

[0026] Video preprocessing unit: used to preprocess the acquired interview video, including video format conversion, resolution adjustment and audio separation;

[0027] Semantic analysis unit: used to perform semantic analysis on the speech transcription content in the interview video using natural language processing;

[0028] Comprehensive scoring unit: used to calculate the interview score of the applicant based on the results of semantic analysis and the preset scoring criteria;

[0029] Score recording unit: used to record interview scores and associate them with the corresponding applicant resumes.

[0030] Preferably, the comprehensive analysis module includes

[0031] Data acquisition unit: used to obtain resume matching data from the resume screening module and interview score data from the interview analysis module;

[0032] Weight preset unit: used to preset the weights of resume matching and interview scores;

[0033] Comprehensive score calculation unit: used to calculate the comprehensive score of each applicant according to the preset weight;

[0034] Threshold screening unit: used to screen out applicants whose comprehensive scores are greater than the preset comprehensive score threshold and generate a list of qualified applicants;

[0035] Sorting unit: used to sort qualified applicants according to their comprehensive scores to form a comprehensive score queue, where the applicants with the highest scores are ranked first;

[0036] Data storage unit: used to store the comprehensive score queue in the system database and associate it with the corresponding applicant resume, resume matching degree, interview score and interview video.

[0037] Preferably, the push module includes

[0038] Comprehensive score queue push unit: used to push the comprehensive score queue generated by the comprehensive analysis module to the recruitment officer;

[0039] Recruiter selection unit: used for recruiters to select candidates based on their comprehensive scores, resume information and interview videos;

[0040] Employment notification sending unit: used to send the employment notification to the selected applicants and record the sending status;

[0041] On-the-job reply unit: After receiving the on-the-job notice, the applicant can reply to confirm whether to accept the invitation and record the applicant's confirmation status;

[0042] The unit for sending the list of new employees is used to integrate the list of new employees according to the status of the applicants accepting the invitation to join the company, and to show the list of new employees to the interviewer.

[0043] The artificial intelligence driven recruitment and interview analysis method provided by the present invention is characterized by comprising the following steps:

[0044] S1: Obtain the resume of the applicant, parse the resume, calculate the matching degree between the resume and the preset job requirements, and screen out the applicants whose resume matching degree is greater than the preset matching degree threshold;

[0045] S2: Send a video interview invitation to the applicant whose resume matching degree is greater than a preset threshold. If the applicant accepts the video interview invitation, he / she can click to enter the interview start page. After the interview video is recorded, the interview video of the applicant is obtained, and the interview video is semantically analyzed to obtain the interview score;

[0046] S3: Generate a comprehensive score based on resume matching and interview scores, screen out applicants whose comprehensive scores are greater than a preset comprehensive score threshold, and form a comprehensive score queue, in which the applicants with the highest scores are ranked first;

[0047] S4: Push the comprehensive score queue to the recruiter. If the recruiter selects a candidate in the comprehensive score queue, the recruitment notice will be sent to the selected candidate.

[0048] Preferably, the degree of match between the resume and the preset job requirements is calculated in S1, specifically, extracting the key job information of the preset job requirements; obtaining the key resume information of the resume parsing unit; preprocessing the key job information and resume information; using a vectorization method to convert the job requirements and resume content into feature vectors; calculating the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain the match degree; normalizing the calculated match degree; and weighting the calculated match degree according to the preset weights of different parts in the job requirements to generate a final match degree.

[0049] The present invention discloses an artificial intelligence-driven recruitment and interview analysis system and method, which has the following beneficial effects: by automatically parsing resumes and calculating the matching degree with job requirements, the time for manual resume screening can be reduced, and candidates who meet the requirements can be quickly locked, the response speed of recruitment can be improved, and the time consumption between recruiters and applicants can be reduced. Through video interview invitations and automatic semantic analysis, the recruitment process is made smoother and more efficient; according to the preset job requirements and matching degree standards, resumes that do not meet the conditions can be excluded, and the problem of low-quality candidate screening caused by personal misjudgment of recruiters can be reduced. Through semantic analysis and scoring of interview videos, the performance of applicants can be objectively evaluated, the interference of personal bias or subjective judgment of interviewers can be avoided, and the accuracy of recruitment decisions can be improved. The resume matching degree and interview score are comprehensively evaluated to form a comprehensive score. This method can more comprehensively and systematically reflect the overall ability of each candidate, avoid the one-sidedness of a single dimension, and improve the accuracy of the final screening. The sorting of the comprehensive score queue helps recruiters quickly identify the best candidates and improves the scientific nature of recruitment decisions. The automated screening, analysis and push functions can significantly reduce the work pressure of recruiters, avoid the heavy manual screening and interview evaluation work, and allow the recruitment team to focus on higher-level judgments in the interview and improve work efficiency. Through semantic analysis of the interview video, the subjective views of the interviewer can be quantified and objective data support can be provided. Such quantitative evaluation not only helps the decision-making of recruiters, but also provides an effective basis for subsequent recruitment optimization and candidate data analysis. Through the technicalization and automation of various links such as resume screening, interview analysis, and comprehensive evaluation, the recruitment process will be more standardized and standardized, which can avoid the problems of inefficiency and information asymmetry common in traditional recruitment. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0051] Figure 1 An architectural diagram of the artificial intelligence-driven recruitment and interview analysis system provided by the present invention;

[0052] Figure 2 A unit structure diagram of a resume screening module in the artificial intelligence driven recruitment and interview analysis system provided by the present invention;

[0053] Figure 3 A subunit diagram of a job requirement matching degree calculation unit in a resume screening module provided by the present invention;

[0054] Figure 4 A unit structure diagram of the interview analysis module in the artificial intelligence driven recruitment and interview analysis system provided by the present invention;

[0055] Figure 5 A unit structure diagram of the comprehensive analysis module in the artificial intelligence driven recruitment and interview analysis system provided by the present invention;

[0056] Figure 6 A unit structure diagram of a push module in the artificial intelligence driven recruitment and interview analysis system provided by the present invention; DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0058] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0059] refer to Figure 1 The artificial intelligence driven recruitment and interview analysis system provided by the present invention includes the following:

[0060] Resume screening module 11: used to obtain the resumes of applicants, analyze the resumes, calculate the matching degree between the resumes and the preset job requirements, and screen out applicants whose resume matching degree is greater than a preset matching degree threshold;

[0061] Interview analysis module 12: used to send video interview invitations to applicants whose resume matching degree is greater than a preset threshold. If the applicant accepts the video interview invitation, a video interview is arranged. After the interview, the interview video of the applicant is obtained, and semantic analysis is performed on the interview video to obtain the interview score;

[0062] Comprehensive analysis module 13: used to generate a comprehensive score based on resume matching and interview scores, screen out applicants whose comprehensive scores are greater than a preset comprehensive score threshold, and form a comprehensive score queue, in which the applicants with the highest scores are ranked first;

[0063] Push module 14: used to push the comprehensive score queue to the recruitment officer. If the recruitment officer selects a candidate in the comprehensive score queue, the employment notice will be sent to the selected candidate.

[0064] The artificial intelligence-driven recruitment and interview analysis system provided by the present invention can automatically parse resumes and calculate the matching degree with job requirements, reduce the time of manual resume screening, quickly lock in candidates who meet the requirements, improve the response speed of recruitment, reduce the time consumption between recruiters and applicants, and make the recruitment process smoother and more efficient through video interview invitations and automatic semantic analysis; according to the preset job requirements and matching degree standards, it can exclude resumes that do not meet the conditions, and reduce the problem of low-quality candidate screening caused by personal misjudgment of recruiters. Through semantic analysis and scoring of interview videos, the performance of applicants is objectively evaluated, the interference of personal bias or subjective judgment of interviewers is avoided, and the accuracy of recruitment decisions is improved. The resume matching degree and interview score are comprehensively evaluated to form a comprehensive score, which reflects the overall ability of each candidate, avoids the one-sidedness of a single dimension, and improves the accuracy of the final screening. The sorting of the comprehensive score queue helps recruiters quickly identify the best candidates and improves the scientific nature of recruitment decisions. The automated screening, analysis and push functions can significantly reduce the work pressure of recruiters, avoid the heavy manual screening and interview evaluation work, and allow the recruitment team to focus on higher-level judgments in interviews and improve work efficiency. Through semantic analysis of interview videos, the subjective opinions of interviewers can be quantified and objective data support can be provided. Such quantitative evaluation not only helps recruiters make decisions, but also provides an effective basis for subsequent recruitment optimization and candidate data analysis. Through the technology and automation of various links such as resume screening, interview analysis, and comprehensive evaluation, the recruitment process will be more standardized and standardized, and can avoid the problems of inefficiency and information asymmetry common in traditional recruitment.

[0065] refer to Figure 2 In a preferred embodiment, the resume screening module 11 includes:

[0066] Resume acquisition unit 111: used to acquire the resume of the applicant, where the resume format includes but is not limited to multiple common formats such as text, PDF, Word, etc.;

[0067] Resume parsing unit 112: used to extract key information from the resume using natural language processing, where the key information in the resume includes personal information, work experience, educational background, skills and expertise, etc.;

[0068] Job requirement matching degree calculation unit 113: used to calculate the matching degree between the applicant's resume and the job requirements according to the preset job requirements;

[0069] Matching degree screening unit 114: used to screen out applicants whose matching degree is greater than a preset matching degree threshold according to the calculated matching degree;

[0070] Data transmission unit 115: used to transmit the corresponding resume and matching degree of the applicant whose matching degree is greater than the preset matching degree threshold to the interview analysis module.

[0071] refer to Figure 3 In a preferred embodiment, the job requirement matching degree calculation unit 113 includes

[0072] Position requirement extraction subunit 1131: used to extract key position information of preset position requirements, including but not limited to position title, position description, skill requirements, work experience requirements, education background requirements, required certificates, etc.;

[0073] Resume information acquisition subunit 1132: used to acquire resume key information from the resume parsing unit;

[0074] Text preprocessing subunit 1133: used to preprocess key information of positions and resumes, including word segmentation, removal of stop words, standardization, etc., to ensure that the data is in a consistent format when performing matching calculations;

[0075] Feature vectorization subunit 1134: used to convert job requirements and resume content into feature vectors using vectorization methods such as TF-IDF, Word2Vec, BERT, etc.;

[0076] Matching degree calculation subunit 1135: used to calculate the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain a matching degree;

[0077] The matching degree normalization processing unit 1136 is used to perform normalization processing on the calculated matching degree;

[0078] Comprehensive matching degree value generating unit 1137: used to perform weighted processing on the calculated matching degrees according to the preset weights of different parts in the job requirements to generate the final matching degree.

[0079] refer to Figure 4 In a preferred embodiment, the interview analysis module 12 includes

[0080] The video interview invitation sending unit 121 is used to send the video interview invitation to the applicant whose resume matching degree is greater than a preset threshold, and record the invitation status;

[0081] Interview video acquisition unit 122: used to acquire the interview video of the applicant through the built-in video recording function after the applicant accepts the video interview invitation;

[0082] Video preprocessing unit 123: used to preprocess the acquired interview video, including video format conversion, resolution adjustment, audio separation, etc., to ensure that the video quality meets the analysis requirements;

[0083] Semantic analysis unit 124: used for semantic analysis of the speech-to-text content in the interview video using natural language processing;

[0084] Specifically, first, identify the speech: used for converting the speech part in the interview video into text using speech recognition technologies such as Google Speech-to-Text, DeepSpeech, or Kaldi to generate speech-to-text content; second, preprocess the converted text: used for performing word segmentation on the speech-to-text content, splitting the text into independent words; performing part-of-speech tagging on the segmented words to identify the part of speech of each word such as nouns, verbs, adjectives, etc.; removing stop words such as "de", "shi", "zai", etc. in the text to reduce interference information. Third, extract the keywords in the text: used for extracting keywords in the interview text using TF-IDF, TextRank, or BERT models, and the keywords include skills and experience, educational background, and job stability, etc.;

[0085] Comprehensive scoring unit 125: used for calculating the interview score of the applicant according to the results of semantic analysis and in combination with the preset scoring criteria;

[0086] Specifically, match the keywords extracted from the applicant's answer content with the preset keywords, and use the BERT model to evaluate the matching degree between the keywords extracted from the answer content and the preset keywords, and calculate the final comprehensive score according to the matching degree.

[0087] Score recording unit 126: used for recording the interview score and associating it with the corresponding applicant's resume.

[0088] Reference Figure 5 In the preferred embodiment, the comprehensive analysis module 13 includes

[0089] Data acquisition unit 131: used for obtaining resume matching degree data from the resume screening module and obtaining interview score data from the interview analysis module;

[0090] Weight preset unit 132: used for presetting the weights of resume matching degree and interview score;

[0091] Comprehensive score calculation unit 133: used for calculating the comprehensive score of each applicant according to the preset weights, and the specific calculation formula is as follows:

[0092] Comprehensive score = (resume matching degree × resume weight) + (interview score × interview weight);

[0093] Threshold screening unit 134: used for screening out applicants with a comprehensive score greater than the preset comprehensive score threshold to generate a list of qualified applicants;

[0094] Sorting unit 135: used to sort qualified applicants according to comprehensive scores to form a comprehensive score queue, in which the applicant with the highest score is at the front;

[0095] Data storage unit 136: used to store the comprehensive score queue in the system database and associate it with the corresponding applicant resume, resume matching degree, interview score and interview video.

[0096] refer to Figure 6 In a preferred embodiment, the push module 14 includes

[0097] Comprehensive score queue push unit 141: used to push the comprehensive score queue generated by the comprehensive analysis module to the recruitment officer; the push method includes but is not limited to email, system notification, recruitment management system interface display, etc.;

[0098] Recruiter selection unit 142: used for recruiters to select applicants based on the applicants' comprehensive scores, resume information and interview videos;

[0099] The employment notice sending unit 143 is used to send the employment notice to the selected applicant and record the sending status. The content of the employment notice includes but is not limited to the employment date, job title, salary, work location, materials required for employment, etc.;

[0100] On-the-job reply unit 144: used for the applicant to reply to confirm whether to accept the on-the-job invitation after receiving the on-the-job notice, and record the confirmation status of the applicant;

[0101] The incoming personnel list sending unit 145 is used to compile an incoming personnel list according to the status of the applicants accepting the job invitation, and present the incoming personnel list to the interviewer.

[0102] The artificial intelligence-driven recruitment and interview analysis method provided by the present invention comprises the following steps:

[0103] S1: Obtain the resume of the applicant, analyze the resume, calculate the matching degree between the resume and the preset job requirements, and screen out the applicants whose resume matching degree is greater than the preset matching degree threshold;

[0104] S2: Send a video interview invitation to the applicant whose resume matching degree is greater than a preset threshold. If the applicant accepts the video interview invitation, he / she can click to enter the interview start page. After the interview video is recorded, the interview video of the applicant is obtained, and the interview video is semantically analyzed to obtain the interview score;

[0105] S3: Generate a comprehensive score based on resume matching and interview scores, screen out applicants whose comprehensive scores are greater than a preset comprehensive score threshold, and form a comprehensive score queue, in which the applicants with the highest scores are ranked first;

[0106] S4: Push the comprehensive score queue to the recruiter. If the recruiter selects a candidate in the comprehensive score queue, the recruitment notice will be sent to the selected candidate.

[0107] In a preferred embodiment, the degree of match between the resume and the preset job requirements is calculated in S1, specifically, extracting the key information of the preset job requirements; obtaining the key information of the resume from the resume parsing unit; preprocessing the key information of the job and the resume information; using a vectorization method to convert the job requirements and the resume content into feature vectors; calculating the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain the match; normalizing the calculated match; and weighting the calculated match according to the preset weights of different parts in the job requirements to generate a final match.

Claims

1. AI-driven recruitment and interview analysis system, characterized by: These include: Resume screening module: used to obtain applicants' resumes, parse them, calculate the matching degree between the resumes and the preset job requirements, and screen out applicants whose resume matching degree is greater than the preset matching degree threshold; Interview analysis module: used to send video interview invitations to candidates whose resume matching degree is greater than the preset threshold. If the candidate accepts the video interview invitation, he / she can click to enter the interview start page. After the interview video is recorded, the interview video of the candidate is obtained and semantic analysis is performed on the interview video to obtain the interview score. Comprehensive analysis module: used to generate comprehensive scores based on resume matching and interview scores, screen out candidates whose comprehensive scores are greater than the preset comprehensive score threshold, and form a comprehensive score queue, in which the candidates with the highest scores are ranked first; Push module: used to push the comprehensive score queue to the recruiter. If the recruiter selects the applicant in the comprehensive score queue, the employment notice will be sent to the selected applicant.

2. The artificial intelligence driven recruitment and interview analysis system according to claim 1, characterized in that: The resume screening module includes: Resume acquisition unit: used to obtain the resume of applicants; Resume parsing unit: used to extract key information from a resume using natural language processing, where the key information in the resume includes personal information, work experience, educational background, and skills and expertise; Job requirement matching degree calculation unit: used to calculate the matching degree between the applicant's resume and the job requirements based on the preset job requirements and key information in the resume; Matching degree screening unit: used to screen out applicants whose matching degree is greater than a preset matching degree threshold according to the calculated matching degree; Data transmission unit: used to transmit the resume and matching degree of the applicant whose matching degree is greater than the preset matching degree threshold to the interview analysis module.

3. The artificial intelligence driven recruitment and interview analysis system according to claim 2, characterized in that: The job requirement matching degree calculation unit includes Position requirement extraction subunit: used to extract key position information required by the preset position; Resume information acquisition subunit: used to acquire resume key information from the resume parsing unit; Text preprocessing subunit: used to preprocess key job information and resume information; Feature vectorization subunit: used to convert job requirements and resume content into feature vectors using vectorization methods; Matching degree calculation subunit: used to calculate the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain the matching degree; Matching degree value normalization processing unit: used for normalizing the calculated matching degree; Comprehensive matching value generation unit: used to perform weighted processing on the calculated matching degree according to the preset weights of different parts of the job requirements to generate the final matching degree.

4. The artificial intelligence driven recruitment and interview analysis system according to claim 1, characterized in that: The interview analysis module includes Video interview invitation sending unit: used to send video interview invitations to applicants whose resume matching degree is greater than a preset threshold, and record the invitation status; Interview video acquisition unit: used to acquire the interview video of the applicant through the built-in video recording function after the applicant accepts the video interview invitation; Video preprocessing unit: used to preprocess the acquired interview video, including video format conversion, resolution adjustment and audio separation; Semantic analysis unit: used to perform semantic analysis on the speech transcription content in the interview video using natural language processing; Comprehensive scoring unit: used to calculate the interview score of the applicant based on the results of semantic analysis and the preset scoring criteria; Score recording unit: used to record interview scores and associate them with the corresponding applicant resumes.

5. The artificial intelligence driven recruitment and interview analysis system according to claim 1, characterized in that: The comprehensive analysis module includes Data acquisition unit: used to obtain resume matching data from the resume screening module and interview score data from the interview analysis module; Weight preset unit: used to preset the weights of resume matching and interview scores; Comprehensive score calculation unit: used to calculate the comprehensive score of each applicant according to the preset weight; Threshold screening unit: used to screen out applicants whose comprehensive scores are greater than the preset comprehensive score threshold and generate a list of qualified applicants; Sorting unit: used to sort qualified applicants according to their comprehensive scores to form a comprehensive score queue, where the applicants with the highest scores are ranked first; Data storage unit: used to store the comprehensive score queue in the system database and associate it with the corresponding applicant resume, resume matching degree, interview score and interview video.

6. The artificial intelligence driven recruitment and interview analysis system according to claim 1, characterized in that: The push module includes Comprehensive score queue push unit: used to push the comprehensive score queue generated by the comprehensive analysis module to the recruitment officer; Recruiter selection unit: used for recruiters to select candidates based on their comprehensive scores, resume information and interview videos; Employment notification sending unit: used to send the employment notification to the selected applicants and record the sending status; Onboarding reply unit: used for applicants to reply to confirm whether they accept the onboarding invitation after receiving the onboarding notice, and record the applicant's confirmation status; The unit for sending the list of new employees is used to integrate the list of new employees according to the status of the applicants accepting the invitation to join the company, and to show the list of new employees to the interviewer.

7. AI-driven recruitment and interview analysis method, characterized by: The steps include: S1: Obtain the resume of the applicant, parse the resume, calculate the matching degree between the resume and the preset job requirements, and screen out the applicants whose resume matching degree is greater than the preset matching degree threshold; S2: Send a video interview invitation to the applicant whose resume matching degree is greater than a preset threshold. If the applicant accepts the video interview invitation, he / she can click to enter the interview start page. After the interview video is recorded, the interview video of the applicant is obtained, and the interview video is semantically analyzed to obtain the interview score; S3: Generate a comprehensive score based on resume matching and interview scores, screen out applicants whose comprehensive scores are greater than a preset comprehensive score threshold, and form a comprehensive score queue, in which the applicants with the highest scores are ranked first; S4: Push the comprehensive score queue to the recruiter. If the recruiter selects a candidate in the comprehensive score queue, the recruitment notice will be sent to the selected candidate.

8. The artificial intelligence driven recruitment and interview analysis method according to claim 7, characterized in that: In the S1, the matching degree between the resume and the preset job requirements is calculated, specifically, extracting the key information of the preset job requirements; obtaining the key information of the resume from the resume parsing unit; preprocessing the key information of the job and the resume information; using a vectorization method to convert the job requirements and the resume content into feature vectors; calculating the similarity between the resume feature vector and the job requirement feature vector by using cosine similarity to obtain the matching degree; normalizing the calculated matching degree; and weighting the calculated matching degree according to the preset weights of different parts in the job requirements to generate a final matching degree.

Images