An intelligent resume system based on AI model

Abstract

The present invention relates to the field of artificial intelligence technology, and specifically to an intelligent resume system based on an AI model, including: an AI writing module generates a resume reference format according to a user's basic information and target position requirements; an AI editing module gives a first modification suggestion; an AI diagnosis module scores the user's resume; and an AI matching module matches the user's resume with the position being recruited using an AI matching model, and recommends positions to the user based on the matching results. The AI ​​writing module of the present invention greatly shortens the time for users to write resumes from scratch by automatically generating a resume reference format. The AI ​​editing model can quickly identify potential problems in resumes and improve the professionalism and readability of resumes. By comprehensively evaluating and scoring resumes through an AI diagnostic model, users can intuitively understand the advantages and disadvantages of resumes and help users identify weak links in resumes. Using the AI ​​matching model, the system can accurately match user resumes with the positions being recruited, improving delivery accuracy and success rate.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent resume system based on an AI model. Background Art

[0002] With the rapid development of artificial intelligence technology, its application in various industries is becoming more and more extensive, especially in the field of job search and recruitment, where the potential of AI technology is being deeply explored. Traditional resume preparation and screening methods often rely on manual operations, which is not only inefficient but also easily affected by subjective judgment, resulting in low matching accuracy. In today's fast-paced information iteration, this traditional method can no longer meet market demand.

[0003] To solve these problems, intelligent resume systems based on AI models have emerged. These systems use natural language processing (NLP) technology to parse and understand the language information in resumes, and use machine learning (ML) algorithms to identify key elements in resumes, such as skills, experience, educational background, etc., and match this information with job descriptions.

[0004] However, the intelligent resume system in the prior art requires users to prepare a resume by themselves and then submit it, which takes a long time to prepare. Summary of the invention

[0005] (I) Purpose of the invention

[0006] The purpose of the present invention is to provide an intelligent resume system based on an AI model that reduces pre-preparation time and improves delivery accuracy.

[0007] (II) Technical solution

[0008] To solve the above problems, the present invention provides an intelligent resume system based on an AI model, comprising:

[0009] AI writing assistance module, AI editing assistance module, AI diagnosis module and AI matching module;

[0010] The AI ​​writing module generates a resume reference format based on the user's basic information and target position requirements;

[0011] The AI ​​modification module uses the AI ​​modification model to give a first modification suggestion to the user resume;

[0012] The AI ​​diagnosis module uses an AI diagnosis model to score the user resume;

[0013] The AI ​​matching module uses an AI matching model to match user resumes with the positions being recruited, and recommends positions to users based on the matching results.

[0014] In another aspect of the present invention, preferably, the AI ​​writing module generates a resume reference format based on the user's basic information and target position requirements, including:

[0015] Using the trained GLM-4 model, a first draft of the resume is generated based on the user's basic information and target position requirements;

[0016] The GLM-4 model is trained using the following loss function:

[0017]

[0018] Among them, L represents the loss function, z s represents the original training sample, z t represents the positive sample of the original training sample, z k represents the negative sample of the original training sample, sim(z s ,z t ) represents the similarity between the original training sample and the positive sample of the original training sample, sim(z s ,z k ) represents the similarity between the original training sample and the negative sample of the original training sample, N represents the number of negative samples of the original training sample, and Γ represents a hyperparameter that adjusts the magnification of the similarity.

[0019] In another aspect of the present invention, preferably, the AI ​​modification module uses the AI ​​modification model to give a first modification suggestion to the user resume, including:

[0020] Segment the Chinese text that does not contain separators to form a plurality of first character strings;

[0021] Matching a plurality of character strings with words existing in the resume corpus, storing unregistered words in a suspected error list, and forming a plurality of second character strings;

[0022] Calculating the confidence corresponding to each first character string by using a preset first algorithm and calculating the window movement score corresponding to each first character string by using a preset second algorithm;

[0023] comparing the confidence level corresponding to the first character string with a first threshold and comparing the window movement score corresponding to the first character string with a second threshold;

[0024] If the confidence score corresponding to the first character string is greater than a first threshold and the window movement score corresponding to the first character string is less than a second threshold, the first character string is an erroneous character string;

[0025] The erroneous character string is marked as a first modification suggestion.

[0026] In another aspect of the present invention, preferably, the calculation formula for calculating the confidence level of each first character string using the preset first algorithm is:

[0027]

[0028] Among them, C(q i ) represents the first string q i The corresponding confidence, freq(q i ) is the first string q i The frequency of occurrence in the corpus, scores(i) represents the first string q i The window moving score, A is the average of all scores, n is the first string q i , α and β represent weight coefficients.

[0029] In another aspect of the present invention, preferably, the calculation formula for calculating the window movement score corresponding to each first character string using the preset second algorithm is:

[0030]

[0031] Among them, P represents the resume corpus, q i represents the first string, x represents the unit length of the window movement, i represents the first string q i The number of exc means if q i The score of appearing in the resume corpus; back_pro(q i-1 ) indicates q i-1 The window moving score is , and U is a constant.

[0032] In another aspect of the present invention, preferably, the AI ​​diagnosis module scores the user resume using the AI ​​diagnosis model, including calculating the comprehensive score using the following formula:

[0033] Y=λ1·V1+λ2·V2+λ3·V3+λ4·V4+λ5·V5

[0034] Among them, Y represents the comprehensive score, λ1 represents the information integrity weight, V1 represents the information integrity score; λ2 represents the aesthetic weight, V2 represents the aesthetic score; λ3 represents the professional skill weight, V3 represents the professional skill score; λ4 represents the work experience weight, V4 represents the work experience score; λ5 represents the personal advantage weight, V5 represents the personal advantage score, λ1+λ2+λ3+λ4+λ5=1.

[0035] In another aspect of the present invention, preferably, the information integrity score is calculated using the following formula:

[0036]

[0037] Among them, V1 represents the information completeness score, ω j represents the weight of the jth information sub-item, η j represents the completeness of the jth information sub-item, and m represents the total number of information sub-items;

[0038] The aesthetic score was calculated using the following formula:

[0039]

[0040] Among them, V2 represents the aesthetic score, G1 represents the neatness of the layout, G2 represents the color matching, G3 represents the font fusion, and G4 represents the image fusion.

[0041] In another aspect of the present invention, preferably,

[0042] The occupational skill score is calculated using the following formula:

[0043]

[0044] Among them, V3 represents the professional skill score, φ r represents the weight of the rth professional skill certificate, ψ r represents the value of the rth vocational skill certificate, and R represents the total number of vocational skill certificates;

[0045] The work experience score is calculated using the following formula:

[0046]

[0047] Among them, V4 represents the work experience score, b1 represents the years of work experience, b2 represents the performance bonus, b3 represents the number of projects, b4 represents the management team score, and b5 represents the number of months in office;

[0048] The personal strengths score is calculated using the following formula:

[0049]

[0050] Among them, V5 represents the personal advantage score, ζ d represents the dth specialty score, and D represents the number of specialties.

[0051] In another aspect of the present invention, preferably, the AI ​​matching module uses an AI matching model to match the user resume with the job being recruited, and recommends a job to the user based on the matching result, including:

[0052] Perform keyword screening on the job positions and the currently recruiting positions in the user's resume, and obtain a list of currently recruiting positions that match the job positions in the user's resume, which is represented as a first list;

[0053] The education level in the user's resume is screened against the education level in the first list, and a list of currently recruiting positions that matches the education level in the user's resume is obtained, which is represented as the second list;

[0054] Screening the professional skill certificates in the user's resume and the professional skill certificates in the second list, obtaining a list of currently recruiting positions that match the professional skill certificates in the user's resume, which is represented as a third list;

[0055] Calculate the matching value between the user's resume and the positions currently being recruited in the third list, and recommend positions to the user based on the matching value from high to low.

[0056] In another aspect of the present invention, preferably, the calculating the matching value between the user resume and the positions currently being recruited in the third list comprises calculating using the following formula:

[0057]

[0058] Where d(E,F) represents the matching value between the user resume and the positions currently being recruited in the third list; E w represents the user resume vector, w represents the number of the user resume vector, h represents the number of user resume vectors, F x represents the vector of the positions currently being recruited in the third list, and x represents the number of the vector of the positions currently being recruited in the third list.

[0059] (III) Beneficial effects

[0060] The above technical solution of the present invention has the following beneficial technical effects:

[0061] The AI ​​writing module of the present invention greatly shortens the time for users to write resumes from scratch by automatically generating resume reference formats. The module can intelligently recommend appropriate resume structures and content highlights based on the user's basic information and target position requirements, helping users quickly build a resume framework that meets industry standards and is highly targeted. After the user initially completes the resume, the AI ​​revision model can quickly identify potential problems in the resume, such as grammatical errors, unclear expressions, irregular formats, etc., and give specific modification suggestions. This instant feedback mechanism helps users quickly optimize the content of their resumes and improve the professionalism and readability of their resumes. Through the AI ​​diagnostic model, the resume is comprehensively evaluated and scored, so that users can intuitively understand the advantages and disadvantages of the resume and help users identify weak links in the resume. Using the AI ​​matching model, the system can accurately match the user's resume with the position being recruited, improving the accuracy and success rate of delivery. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the overall structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0064] Obviously, the described embodiments are only some embodiments of the present invention, but not all 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.

[0065] In the description of the present invention, it should be noted that the terms “first”, “second” and “third” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0066] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0067] Embodiment 1

[0068] An intelligent resume system based on AI model, Figure 1 FIG. 1 shows a schematic diagram of the overall structure of an embodiment of the present invention. Figure 1 As shown, including:

[0069] AI writing assistance module, AI editing assistance module, AI diagnosis module and AI matching module;

[0070] The AI ​​writing module generates a resume reference format based on the user's basic information and the target position requirements; the user enters his or her basic information, such as name, contact information, educational background, work experience, etc., and specifies the requirements of the target position. The system combines a large number of industry standards and samples of excellent resumes to intelligently generate a targeted resume reference format; in this embodiment, the AI ​​writing module generates a first draft of the resume based on the user's basic information and the target position requirements, including:

[0071] Using the trained GLM-4 model, a first draft of the resume is generated based on the user's basic information and target position requirements;

[0072] The GLM-4 model is trained using the following loss function:

[0073]

[0074] Among them, L represents the loss function, z s represents the original training sample, z t represents the positive sample of the original training sample, z k represents the negative sample of the original training sample, sim(z s ,z t ) represents the similarity between the original training sample and the positive sample of the original training sample, sim(z s ,z k ) represents the similarity between the original training sample and the negative sample of the original training sample, N represents the number of negative samples of the original training sample, and Γ represents a hyperparameter that adjusts the magnification of the similarity. This loss function optimizes the model by maximizing the similarity between positive samples and minimizing the similarity between negative samples, which helps the model learn more discriminative feature representations in complex data sets, thereby improving the performance of the model.

[0075] The AI ​​editing module uses the AI ​​editing model to give a first modification suggestion to the user resume; the resume source of the AI ​​editing module can be the AI ​​writing module, or it can be a resume uploaded by the user, or it can be a resume rewritten according to the reference format of the AI ​​writing module; the specific content of the first modification suggestion is not limited here, and it can optionally include grammatical errors, spelling errors, repetitions and other problems. In this embodiment, the AI ​​editing module uses the AI ​​editing model to give a first modification suggestion to the user resume, including:

[0076] Segment the Chinese text that does not contain separators to form a plurality of first character strings; the specific method of segmentation is not limited here, and optionally, a rule-based segmentation method, such as maximum forward matching (MM), maximum reverse matching (RMM), etc., is used for segmentation; a plurality of first character strings, i.e., words or phrases after segmentation;

[0077] Match several character strings with words in the resume corpus, store unregistered words in a suspected error list, and form several second character strings; traverse all first character strings and compare them with words in the resume corpus one by one. If a first character string does not find a completely matched word in the corpus, it is regarded as an unregistered word and put into the suspected error list;

[0078] The confidence corresponding to each first character string is calculated using a preset first algorithm, and the window movement score corresponding to each first character string is calculated using a preset second algorithm. In this embodiment, the confidence corresponding to each first character string calculated using the preset first algorithm includes calculating the confidence using the following formula:

[0079] in,

[0080] Among them, C(q i ) represents the first string q i The corresponding confidence, freq(q i ) is the first string q i The frequency of occurrence in the corpus, scores(i) represents the first string q i The window moving score, A is the average of all scores, n is the first string q i , α and β represent weight coefficients.

[0081] The using of the preset second algorithm to calculate the window movement score corresponding to each first character string includes using the following formula to calculate the window movement score:

[0082]

[0083] Among them, P represents the resume corpus, q i represents the first string, x represents the unit length of the window movement, i represents the first string q i The number of exc means if q i The score of appearing in the resume corpus; back_pro(q i-1 ) indicates q i-1 The window moving score is , and U is a constant.

[0084] Compare the confidence corresponding to the first character string with a first threshold and compare the window movement score corresponding to the first character string with a second threshold; the specific content of the first threshold is not limited here, and it is optionally 1.5; the content of the second threshold is also not limited, and it is optionally the median of all scores.

[0085] If the confidence score corresponding to the first character string is greater than a first threshold and the window movement score corresponding to the first character string is less than a second threshold, the first character string is an erroneous character string;

[0086] The erroneous character string is marked as a first modification suggestion.

[0087] The AI ​​diagnosis module scores the user resume using the AI ​​diagnosis model; the source of the user resume here can be the reference format of the AI ​​writing module, or it can be the resume uploaded by the user, or it can be the resume rewritten according to the first modification suggestion of the AI ​​modification module; optionally, a scoring report can be given according to the score, and a second modification suggestion can be given;

[0088] The specific content of the scoring is not limited here. In this embodiment, the AI ​​diagnosis module scores the user resume using the AI ​​diagnosis model, including calculating the comprehensive score using the following formula:

[0089] Y=λ1·V1+λ2·V2+λ3·V3+λ4·V4+λ5·V5

[0090] Among them, Y represents the comprehensive score, λ1 represents the information integrity weight, V1 represents the information integrity score; λ2 represents the aesthetic weight, V2 represents the aesthetic score; λ3 represents the professional skill weight, V3 represents the professional skill score; λ4 represents the work experience weight, V4 represents the work experience score; λ5 represents the personal advantage weight, V5 represents the personal advantage score, λ1+λ2+λ3+λ4+λ5=1.

[0091] The information completeness score measures whether the resume contains all necessary and relevant personal information, education background, work experience, skill certificates and other basic information. This information is the part that the employer pays most attention to when initially screening resumes. In this embodiment,

[0092] The information completeness score is calculated using the following formula:

[0093]

[0094] Among them, V1 represents the information completeness score, ω j represents the weight of the jth information sub-item, η j represents the completeness of the jth information sub-item, and m represents the total number of information sub-items;

[0095] The aesthetic score focuses on the quality of the resume's layout, font selection, color matching, and other visual presentation aspects. A beautiful and neat resume can leave a good first impression on the recruiter. In this embodiment,

[0096] The aesthetic score was calculated using the following formula:

[0097]

[0098] Among them, V2 represents the aesthetic score, G1 represents the neatness of the layout, G2 represents the color matching, G3 represents the font fusion, and G4 represents the image fusion.

[0099] The professional skills score measures the skill level and ability related to the job position shown in the resume. These skills should be specific, quantifiable, and highly matched to the job requirements. In this embodiment, the professional skills certificate is used for calculation.

[0100] The occupational skill score is calculated using the following formula:

[0101]

[0102] Among them, V3 represents the professional skill score, φ r represents the weight of the rth professional skill certificate, ψ r represents the value of the rth vocational skill certificate, and R represents the total number of vocational skill certificates;

[0103] The work experience score focuses on whether the work experience described in the resume is relevant to the target position and whether the length and quality of the work experience meet the requirements. Work experience is an important basis for evaluating whether an applicant is qualified for a position. In this embodiment,

[0104] The work experience score is calculated using the following formula:

[0105]

[0106] Among them, V4 represents the work experience score, b1 represents the years of work experience, b2 represents the performance bonus, b3 represents the number of projects, b4 represents the management team score, and b5 represents the number of months in office;

[0107] The personal strengths score measures the unique strengths, specialties and potential of the applicant as shown in the resume. These strengths can be in terms of skills, personality traits, leadership, teamwork ability, etc. In this embodiment, the personal strengths score is calculated using the following formula:

[0108]

[0109] Among them, V5 represents the personal advantage score, ζ d represents the dth specialty score, and D represents the number of specialties.

[0110] Weight allocation is a key part of the comprehensive scoring formula, which determines the contribution of each scoring dimension to the final score. The weight setting should be flexibly adjusted according to the specific needs of the position and the preferences of the recruiter. For example, for technical positions, the weight of professional skills and work experience may be relatively high; while for positions that focus on teamwork and communication, the weight of personal strengths and work experience may be higher.

[0111] The AI ​​matching module uses an AI matching model to match user resumes with the positions being recruited, and recommends positions to users based on the matching results, including:

[0112] Perform keyword screening on the job positions and the currently recruiting positions in the user's resume, and obtain a list of currently recruiting positions that match the job positions in the user's resume, which is represented as a first list;

[0113] The education level in the user's resume is screened against the education level in the first list, and a list of currently recruiting positions that matches the education level in the user's resume is obtained, which is represented as the second list;

[0114] Screening the professional skill certificates in the user's resume and the professional skill certificates in the second list, obtaining a list of currently recruiting positions that match the professional skill certificates in the user's resume, which is represented as a third list;

[0115] Calculate the matching value between the user's resume and the positions currently being recruited in the third list, and recommend positions to the user based on the matching value from high to low. Through multiple rounds of screening, job keywords, education, professional skills certificates, etc., the AI ​​matching module can gradually narrow the matching range, and the positions in the third list obtained in the end are more compatible with the user's resume. This refined matching method ensures the accuracy and pertinence of the recommended positions.

[0116] The calculation of the matching value between the user resume and the positions currently being recruited in the third list includes using the following formula for calculation:

[0117]

[0118] Where d(E,F) represents the matching value between the user resume and the positions currently being recruited in the third list; E w represents the user resume vector, w represents the number of the user resume vector, h represents the number of user resume vectors, F x represents the vector of the positions currently being recruited in the third list, and x represents the number of the vector of the positions currently being recruited in the third list.

[0119] The AI ​​writing module of the present invention greatly shortens the time for users to write resumes from scratch by automatically generating resume reference formats. The module can intelligently recommend appropriate resume structures and content highlights based on the user's basic information and target position requirements, helping users quickly build a resume framework that meets industry standards and is highly targeted. After the user initially completes the resume, the AI ​​revision model can quickly identify potential problems in the resume, such as grammatical errors, unclear expressions, irregular formats, etc., and give specific modification suggestions. This instant feedback mechanism helps users quickly optimize the content of their resumes and improve the professionalism and readability of their resumes. Through the AI ​​diagnostic model, the resume is comprehensively evaluated and scored, so that users can intuitively understand the advantages and disadvantages of the resume and help users identify weak links in the resume. Using the AI ​​matching model, the system can accurately match the user's resume with the position being recruited, improving the accuracy and success rate of delivery.

[0120] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

[0121] The present invention has been described above with reference to the embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents. Without departing from the scope of the present invention, a person skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

[0122] Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

[0123] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. An intelligent resume system based on AI model, characterized in that: include: AI writing assistance module, AI editing assistance module, AI diagnosis module and AI matching module; The AI ​​writing module generates a resume reference format based on the user's basic information and target position requirements; The AI ​​modification module uses the AI ​​modification model to give a first modification suggestion to the user resume; The AI ​​diagnosis module uses an AI diagnosis model to score the user resume; The AI ​​matching module uses an AI matching model to match user resumes with the positions being recruited, and recommends positions to users based on the matching results; The AI ​​modification module uses the AI ​​modification model to give the first modification suggestion to the user resume, including: Segmenting the Chinese text that does not contain separators to form a plurality of first character strings, wherein the segmentation uses a maximum forward matching method or a maximum reverse matching method; Matching a plurality of character strings with words existing in the resume corpus, storing unregistered words in a suspected error list, and forming a plurality of second character strings; Calculating the confidence corresponding to each first character string by using a preset first algorithm and calculating the window movement score corresponding to each first character string by using a preset second algorithm; comparing the confidence level corresponding to the first character string with a first threshold and comparing the window movement score corresponding to the first character string with a second threshold; If the confidence score corresponding to the first character string is greater than a first threshold and the window movement score corresponding to the first character string is less than a second threshold, the first character string is an erroneous character string; Marking the erroneous character string as a first modification suggestion; The calculation formula for calculating the confidence level of each first character string using the preset first algorithm is: Among them, C(q i ) represents the first string q i The corresponding confidence, freq(q i ) is the first string q i The frequency of occurrence in the corpus, scores(i) represents the first string q i The window moving score, A is the average of all scores, n is the first string q i The number of, α and β represent weight coefficients; The calculation formula for calculating the window movement score corresponding to each first character string using the preset second algorithm is: Among them, P represents the resume corpus, q i represents the first string, x represents the unit length of the window movement, i represents the first string q i The number of exc means if q i The score of appearing in the resume corpus; back_pro(q i-1 ) indicates q i-1 The window moving score is , and U is a constant.

2. The intelligent resume system according to claim 1, characterized in that: The AI ​​writing module generates a resume reference format based on the user's basic information and target position requirements, including: Using the trained GLM-4 model, a first draft of the resume is generated based on the user's basic information and target position requirements; The GLM-4 model is trained using the following loss function: Among them, L represents the loss function, z s represents the original training sample, z t represents the positive sample of the original training sample, z k represents the negative sample of the original training sample, sim(z s ,z t ) represents the similarity between the original training sample and the positive sample of the original training sample, sim(z s ,z k ) represents the similarity between the original training sample and the negative sample of the original training sample, N represents the number of negative samples of the original training sample, and Γ represents a hyperparameter that adjusts the magnification of the similarity.

3. The intelligent resume system according to claim 1, characterized in that: The AI ​​diagnosis module uses the AI ​​diagnosis model to score the user resume, including calculating the comprehensive score using the following formula: Y=λ1·V1+λ2·V2+λ3·V3+λ4·V4+λ5·V5 Among them, Y represents the comprehensive score, λ1 represents the information integrity weight, V1 represents the information integrity score; λ2 represents the aesthetic weight, V2 represents the aesthetic score; λ3 represents the professional skill weight, V3 represents the professional skill score; λ4 represents the work experience weight, V4 represents the work experience score; λ5 represents the personal advantage weight, V5 represents the personal advantage score, λ1+λ2+λ3+λ4+λ5=1.

4. The intelligent resume system according to claim 3, characterized in that: The information completeness score is calculated using the following formula: Among them, V1 represents the information completeness score, ω j represents the weight of the jth information sub-item, η j represents the completeness of the jth information sub-item, and m represents the total number of information sub-items; The aesthetic score was calculated using the following formula: Among them, V2 represents the aesthetic score, G1 represents the neatness of the layout, G2 represents the color matching, G3 represents the font fusion, and G4 represents the image fusion.

5. The intelligent resume system according to claim 4, characterized in that: The occupational skill score is calculated using the following formula: Among them, V3 represents the professional skill score, φ r represents the weight of the rth professional skill certificate, ψ r represents the value of the rth vocational skill certificate, and R represents the total number of vocational skill certificates; The work experience score is calculated using the following formula: Among them, V4 represents the work experience score, b1 represents the years of work experience, b2 represents the performance bonus, b3 represents the number of projects, b4 represents the management team score, and b5 represents the number of months in office; The personal strengths score is calculated using the following formula: Among them, V5 represents the personal advantage score, ζ d represents the dth specialty score, and D represents the number of specialties.

6. The intelligent resume system according to claim 5, characterized in that: The AI ​​matching module uses an AI matching model to match user resumes with the positions being recruited, and recommends positions to users based on the matching results, including: Perform keyword screening on the job positions and the currently recruiting positions in the user's resume, and obtain a list of currently recruiting positions that match the job positions in the user's resume, which is represented as a first list; The education level in the user's resume is screened against the education level in the first list, and a list of currently recruiting positions that matches the education level in the user's resume is obtained, which is represented as the second list; Screening the professional skill certificates in the user's resume and the professional skill certificates in the second list, obtaining a list of currently recruiting positions that match the professional skill certificates in the user's resume, which is represented as a third list; Calculate the matching value between the user's resume and the positions currently being recruited in the third list, and recommend positions to the user based on the matching value from high to low.

7. The intelligent resume system according to claim 6, characterized in that: The calculation of the matching value between the user resume and the positions currently being recruited in the third list includes using the following formula for calculation: Where d(E,F) represents the matching value between the user resume and the positions currently being recruited in the third list; E w represents the user resume vector, h represents the number of user resume vectors, F x A vector representing the positions currently being recruited in the third list.

Images