Method and system for diagnosing ability gap based on professional knowledge graph and large model enhancement

Abstract

The application relates to a capability gap diagnosis method and system based on a professional knowledge graph and a large model enhancement. The method comprises the following steps: constructing a professional capability knowledge graph; determining the mastery degree of a target student on each work skill based on the knowledge graph, and arranging the student individual capability vector; obtaining a target post, calculating the demand intensity of the work skill for the target post, and arranging the post capability demand vector; comparing the mastery degree and the demand intensity on the same skill dimension, taking the skill with a mastery degree lower than the demand intensity as the capability gap, calculating the overall matching degree, and obtaining a capability gap diagnosis result; based on the diagnosis result, obtaining structured knowledge associated with the capability gap from the knowledge graph as a constraint condition to input a large language model, and generating a growth path planning suggestion. The method can realize accurate quantitative diagnosis of student capability and post demand and traceable personalized growth path planning.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence, and in particular relates to a method and system for diagnosing ability gaps based on occupational knowledge graphs and large model enhancement. Background Technology

[0002] With the development of artificial intelligence technology, knowledge graphs and large language models have demonstrated strong knowledge representation and text generation capabilities in vertical fields, and are gradually being applied to scenarios such as career guidance and talent evaluation. However, most existing applications use them as independent tools, with knowledge graphs mainly used for information retrieval and large models mainly used for question-answer generation, lacking deep integration and failing to form a reliable technical link from knowledge modeling to intelligent generation.

[0003] Traditional vocational competency diagnosis mainly relies on qualitative assessments based on human experience, standardized questionnaires, and job information aggregation. Some systems use keyword matching or rule templates to simply compare student data with job requirements. A few solutions directly utilize large language models to generate career suggestions. The basic approach is to assemble student information and target job information into input prompts, and then the model outputs competency analysis and improvement suggestions in one go.

[0004] However, the above methods have significant technical drawbacks: First, there is a lack of structured semantic association and quantitative mapping mechanism between training elements such as course learning and certificate acquisition and job competency requirements, making diagnostic results dependent on subjective experience and difficult to reproduce; second, the large-scale model generation process lacks traceable structured knowledge constraints, and the output content comes from model statistical regularities rather than verifiable data sources, resulting in generalized results and difficulty in tracing responsibility. Based on this, this invention provides a competency gap diagnosis method and system based on occupational knowledge graphs and large-scale model enhancement, aiming to solve the above two core problems. Summary of the Invention

[0005] Based on this, it is necessary to provide a capability gap diagnosis method and system based on occupational knowledge graphs and large models to address the above-mentioned technical problems. This method and system can perform structured modeling and quantitative representation of students' abilities and job requirements based on occupational knowledge graphs, accurately compare individual student ability vectors with job ability requirement vectors to identify capability gaps, and use the structured knowledge in the knowledge graph to constrain the large language model to generate traceable growth path planning suggestions.

[0006] Firstly, this application provides a capability gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement, including:

[0007] S1. Obtain the elements of student ability development, the elements of job ability requirements, and the relationship between the elements of student ability development and the elements of job ability requirements, and construct a professional ability knowledge graph based on the elements of student ability development, the elements of job ability requirements, and the relationship.

[0008] S2. Based on the professional competence knowledge graph, obtain the target student's personal data and work skills, determine the target student's mastery of each work skill based on the personal data, arrange the mastery level into a vector form, and obtain the student's individual competence vector.

[0009] S3. Obtain the target position specified by the user, calculate the demand intensity of the job skills for the target position based on the correlation information between the target position and the job skills, arrange the demand intensity into a vector form, and obtain the job capability demand vector.

[0010] S4. Based on the individual student ability vector and the job ability requirement vector, compare the mastery level and demand intensity on the same skill dimension. Take the job skills with a mastery level lower than the demand intensity as the ability gap of the target student, and calculate the overall matching degree between the target student and the target job. Integrate the ability gap and the overall matching degree to obtain the ability gap diagnosis result.

[0011] S5. Based on the diagnostic results of the ability gap, obtain structured knowledge related to the ability gap from the professional ability knowledge graph, input the structured knowledge as a constraint into the large language model, and generate growth path planning suggestions for the ability gap within the scope of the structured knowledge.

[0012] In one embodiment, based on a professional competence knowledge graph, the personal data and job skills of the target student are obtained. The student's mastery level in each job skill is determined based on the personal data, and the mastery levels are arranged into a vector form to obtain the student's individual competence vector, including:

[0013] S11. Based on the supporting relationship between courses and job skills in the professional competence knowledge graph, as well as the content of courses and the description of job skills, calculate the contribution coefficient of each course to each job skill.

[0014] S12. Obtain the grade data of the courses already taken by the target student, calculate the target student's mastery of each course based on the grade data, and weight and integrate the mastery level and contribution coefficient to obtain the target student's initial mastery of each job skill.

[0015] S13. Obtain the certificate information and practical experience data of the target student. Based on the proof relationship between certificates and work skills, the manifestation relationship between practical experience data and work skills, and the certificate information and practical experience data in the professional ability knowledge graph, calculate the enhancement coefficient of certificates and practical experience on work skills. Combine the initial mastery and enhancement coefficient to obtain the final mastery of each work skill of the target student.

[0016] S14. Arrange all job skills in the professional competence knowledge graph in a fixed order as a dimension index, and input the final mastery of each job skill of the target student into the position of the corresponding skill in the dimension index in sequence to generate the student's individual competence vector.

[0017] In one embodiment, the initial mastery level and the enhancement coefficient are combined to obtain the calculation expression for the target student's final mastery level of each job skill:

[0018] ;

[0019] in, For target students, For job skills, for right The ultimate degree of control, for right Initial level of mastery for The certificate information obtained for Data on practical experience To prove the relationship, To reflect the relationship, For certificate information, Data based on practical experience for right The enhancement coefficient, for right The enhancement coefficient.

[0020] In one embodiment, based on the student's individual ability vector and the job's ability requirement vector, the degree of mastery and the intensity of demand for the same skill dimension are compared. Job skills where the degree of mastery is lower than the intensity of demand are identified as the ability gap of the target student. The overall matching degree between the target student and the target job is calculated. Integrating the ability gap and the overall matching degree, an ability gap diagnosis result is obtained, including:

[0021] S21. Align the student's individual ability vector and the job ability requirement vector by dimension, and calculate the difference between the requirement intensity and the mastery level for each skill dimension to obtain the difference results for each skill dimension.

[0022] S22. Arrange the difference results according to the dimension index to generate a difference vector;

[0023] S23. Based on the difference vector, mark the work skills with a difference greater than zero in each skill dimension as competency gap skills, and take the difference corresponding to each competency gap skill as the competency gap value of the work skill. Summarize the competency gap skills and competency gap values ​​to obtain the competency gap set.

[0024] S24. Based on the individual student ability vector and the job ability requirement vector, calculate the overall matching degree between the target student and the target job according to the numerical relationship between the mastery level and the demand intensity of each skill dimension.

[0025] S25. Based on the capability gap value, classify the severity level of each capability gap skill, and integrate the capability gap set, the severity level of the gap, and the overall matching degree into a structured capability gap diagnosis result.

[0026] In one embodiment, based on the ability gap diagnosis results, structured knowledge related to the ability gap is obtained from a professional ability knowledge graph. This structured knowledge is then input into a large language model as a constraint. Within the scope of this structured knowledge, growth path planning suggestions for the ability gap are generated, including:

[0027] S31. Extract the feature information of each skill gap from the capability gap diagnosis results, and generate knowledge retrieval conditions based on the feature information; wherein, the knowledge retrieval conditions include a first retrieval condition with each skill gap as the retrieval subject and a second retrieval condition with the target position as the retrieval subject;

[0028] S32. Using the first and second retrieval conditions in the knowledge retrieval conditions as inputs for multi-way queries, semantic retrieval is performed and knowledge fragments that semantically match the knowledge retrieval conditions are recalled to obtain a recall result set.

[0029] S33. Using the skills gaps and target positions as query objects, extract the entities and relationships associated with the skills gaps and the target positions from the professional competence knowledge graph to obtain the entity and relationship sequence, and transform the entity and relationship sequence into a structured knowledge context.

[0030] S34. Merge each recall result set and structured knowledge context into a trusted knowledge context. Use the trusted knowledge context as the basis for generation and input it into the large language model to generate growth path planning suggestions within the scope of the trusted knowledge context.

[0031] Secondly, this application also provides a capability gap diagnosis system based on occupational knowledge graphs and large-scale model enhancement, used to implement the method described in the first aspect, including:

[0032] The knowledge graph construction module is used to obtain the elements of student ability development, the elements of job ability requirements, and the relationships between the elements of student ability development and the elements of job ability requirements, and to construct a professional ability knowledge graph based on the elements of student ability development, the elements of job ability requirements, and the relationships.

[0033] The student competency modeling module is used to acquire the personal data and work skills of target students based on the professional competency knowledge graph, determine the degree of mastery of each work skill of the target students based on the personal data, and arrange the degree of mastery into a vector form to obtain the individual competency vector of the student.

[0034] The job requirement modeling module is used to obtain the target job specified by the user, calculate the demand intensity of the job skills for the target job based on the correlation information between the target job and the job skills, and arrange the demand intensity into a vector form to obtain the job capability requirement vector.

[0035] The competency gap diagnosis module is used to compare the mastery level and demand intensity of the same skill dimension based on the individual student's competency vector and the job competency demand vector. The module identifies the job skills where the mastery level is lower than the demand intensity as the competency gap of the target student, calculates the overall matching degree between the target student and the target job, and integrates the competency gap and the overall matching degree to obtain the competency gap diagnosis result.

[0036] The growth path planning module is used to extract structured knowledge related to the ability gap from the professional ability knowledge graph based on the ability gap diagnosis results. The structured knowledge is then input into the large language model as a constraint, and growth path planning suggestions for the ability gap are generated within the scope of the structured knowledge.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the methods in the first aspect of this application.

[0038] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of any of the methods in the first aspect of this application.

[0039] The aforementioned competency gap diagnosis method and system based on occupational knowledge graphs and large-scale model enhancement explicitly and structurally establishes the relationship between student training elements and job requirement elements by constructing an occupational competency knowledge graph. This provides a computable knowledge foundation for subsequent quantitative comparisons. Simultaneously, it maps student training information such as coursework, certification acquisition, and practical experience to the same job skill requirements dimension, eliminating the lack of structured semantic associations and quantitative mapping mechanisms between training elements and job requirements. This transforms competency diagnosis from subjective experience-based judgment to quantifiable vector comparison. Furthermore, it compares individual student competency vectors dimension by dimension with… The difference in mastery between job competency requirement vectors accurately identifies specific competency gaps in students across various job skills and quantifies the degree of these gaps. Simultaneously, it calculates the overall matching degree, ensuring the diagnostic results are quantifiable and reproducible. Finally, by extracting structured knowledge related to competency gaps from a knowledge graph as a reliable context, it generates career path planning suggestions. The traceable entities and relationships in the knowledge graph form knowledge constraints on the generation process of the large model, making the generated content verifiable and traceable. This solves the problems of output generalization and difficulty in accountability caused by the lack of structured knowledge constraints when existing large models directly generate career suggestions. Attached Figure Description

[0040] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of a capability gap diagnosis method based on occupational knowledge graph and large model enhancement in one embodiment of the present invention;

[0042] Figure 2 This is a flowchart of a method for obtaining capability gap diagnosis results in one embodiment of the present invention;

[0043] Figure 3 This is a schematic diagram of the capability gap diagnosis system based on occupational knowledge graph and large model enhancement in one embodiment of the present invention. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0045] In one embodiment, such as Figure 1 As shown, a capability gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement is provided. This embodiment illustrates the application of this method to a diagnostic terminal. It is understood that this method can also be applied to a server, and further to a system including both a diagnostic terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0046] S1. Obtain the elements of student ability development, the elements of job ability requirements, and the relationship between the elements of student ability development and the elements of job ability requirements, and construct a professional ability knowledge graph based on the elements of student ability development, the elements of job ability requirements, and the relationship.

[0047] Specifically, the professional competence knowledge graph can be a structured knowledge model designed independently for the diagnosis of professional competence gaps. Unlike conventional general talent knowledge graphs and general job knowledge graphs, the professional competence knowledge graph can be composed of student competence development element entities, job competence requirement element entities, and semantic relationships between student competence development element entities and job competence requirement element entities. It integrates university talent training standards, enterprise job performance standards and professional skills industry standards, and has the unique characteristics of entity standardization, quantitative relationship and traceable knowledge data, which can be adapted to the competence quantitative diagnosis logic of this invention.

[0048] For example, the diagnostic terminal can complete the standardized modeling of core entities, defining two major core entity categories: student ability development elements and job ability requirement elements. For student ability development elements, it uniformly incorporates various talent training dimensions such as course learning, certificate acquisition, practical training competitions, and internship experience, and configures standardized attribute fields to record core training data. For job ability requirement elements, it aligns with the National Occupational Classification and real job requirements of enterprises, covering core job performance dimensions such as job theoretical knowledge, practical skills, and comprehensive qualities. Based on this, the diagnostic terminal can build a standardized semantic association system between student ability development elements and job ability requirement elements, constructing four types of association relationships: direct matching association, indirect support association, ability complementary association, and ability substitution association. The diagnostic terminal can perform multi-source knowledge fusion and data cleaning, eliminating redundant and repetitive entities and association relationships, and supplementing missing association weight data through large-scale model semantic calculation combined with industry calibration rules. The final structured knowledge is then solidified and stored in a graph database, completing the construction of a professional ability knowledge graph adapted to the ability diagnosis scenario of this invention.

[0049] S2. Based on the professional competence knowledge graph, obtain the target student's personal data and work skills, determine the target student's mastery of each work skill based on the personal data, arrange the mastery level into a vector form, and obtain the student's individual competence vector.

[0050] Specifically, a student's individual ability vector can be a one-dimensional quantitative data sequence formed by uniformly ranking the students' mastery of various work skills based on the standardized skill dimensions of the vocational ability knowledge graph. It has the characteristics of fixed dimensions, unified standards, and vector operation capability. It is the core carrier for realizing the digitization and quantitative comparison of students' abilities. The general student ability evaluation system cannot form this standardized vector data structure.

[0051] For example, the diagnostic terminal can acquire comprehensive training data of the target student and extract effective skill information from unstructured texts such as student growth records and practical training summaries. After data collection, the diagnostic terminal can perform standardized preprocessing operations to unify various data format standards, remove abnormal and invalid data, and filter redundant information without substantial skill support. The diagnostic terminal can map the preprocessed student data to the standardized work skill dimensions built into the knowledge graph through the entity association logic, matching all work skill items corresponding to the student. The diagnostic terminal can calculate the student's mastery score of work skills using a multi-factor weighted scoring algorithm. The expression for calculating the student's mastery score can be:

[0052]

[0053] in, For the first Student proficiency scores for each job skill For students, the corresponding job skills Student development element scores For the first The weight coefficients corresponding to each training element The total number of training elements corresponding to work skills; the diagnostic terminal can normalize all mastery scores to a numerical range of 0 to 1, and finally arrange the quantitative mastery of each skill in order according to the standardized skill dimension sorting rules preset by the knowledge graph, so as to obtain the individual ability vector of the student.

[0054] S3. Obtain the target position specified by the user, calculate the demand intensity of the job skills for the target position based on the correlation information between the target position and the job skills, arrange the demand intensity into a vector form, and obtain the job capability demand vector.

[0055] Specifically, the job competency requirement vector can be a one-dimensional numerical sequence formed by uniformly ranking the demand intensity of various skills for the target job based on the standardized work skill dimension of the professional competency knowledge graph, which is completely aligned with the individual competency vector dimension of the student.

[0056] For example, the diagnostic terminal can locate the entity data corresponding to the target position in the knowledge graph and obtain all the job skill association information bound to the target position. The diagnostic terminal can hierarchically quantify and assign values ​​to the demand intensity of various job skills through three core dimensions: the degree of necessity for job performance, industry common standards, and the contribution of skills to the job position. It can distinguish between core essential skills, important supporting skills, basic auxiliary skills, and irrelevant skills, and complete the weight adjustment in combination with industry standards, so as to unify and normalize the demand intensity of all skills to a numerical range of 0 to 1. The diagnostic terminal can follow the standardized skill dimension sorting order of the student's individual ability vector, arrange the quantified job skill demand intensity values ​​in sequence, and construct the job ability demand vector.

[0057] S4. Based on the individual student ability vector and the job ability requirement vector, compare the mastery level and demand intensity on the same skill dimension. Take the job skills with a mastery level lower than the demand intensity as the ability gap of the target student, and calculate the overall matching degree between the target student and the target job. Integrate the ability gap and the overall matching degree to obtain the ability gap diagnosis result.

[0058] Specifically, the diagnostic results of the competency gap can be a quantitative integrated result that combines single-dimensional skills gaps, the severity of skills gaps, the overall match between students and positions, and individual competency strengths.

[0059] For example, the diagnostic terminal can perform a dimensional skill matching comparison, calculating the difference between the student's mastery level and the job requirement intensity for each set of corresponding dimension skills; the diagnostic terminal can identify skills whose mastery level is lower than the requirement intensity as gap skills, record the ability gap values ​​for each dimension, and summarize them to obtain a quantitative ability gap list; the diagnostic terminal can calculate the overall matching degree between the student and the target job using a weighted Euclidean distance algorithm; wherein, the expression for calculating the overall matching degree can be:

[0060]

[0061] in, To assess the overall match between students and target positions, For the first The job requirement weight corresponding to each job skill For the first The job requirement intensity value for each skill. For the first The numerical value of students' mastery of the job skills. The total number of dimensions for standardized work skills; the diagnostic terminal can constrain the overall matching degree value within the range of 0 to 1, and combine the single-dimensional capability gap list, skill gap classification, advantageous skill dimensions and overall matching degree data to obtain capability gap diagnostic results.

[0062] S5. Based on the diagnostic results of the ability gap, obtain structured knowledge related to the ability gap from the professional ability knowledge graph, input the structured knowledge as a constraint into the large language model, and generate growth path planning suggestions for the ability gap within the scope of the structured knowledge.

[0063] Specifically, growth path planning suggestions can be personalized improvement plans generated by combining knowledge graph constraints and large model enhancement technology; growth path planning suggestions can be layered and implementable plans generated based on students' personalized ability gap data, target job skill requirement standards, and structured related knowledge in knowledge graphs.

[0064] For example, the diagnostic terminal, based on the quantitative results of competency gap diagnosis, combines the structured knowledge of the vocational competency knowledge graph with the capabilities of the large model to generate personalized growth path planning suggestions tailored to students and target positions. The diagnostic terminal can retrieve structured knowledge from the knowledge graph, including job standards, learning resources, training systems, qualification specifications, and skill collaboration relationships corresponding to each skill gap, using reliable graph data to constrain the large model generation process. The diagnostic terminal can calculate the priority weight for skill improvement using a priority weight calculation method; the expression for calculating the priority weight can be:

[0065]

[0066] in, For the first Prioritize and weight the improvement of specific skills. For the first The absolute ability gap value for each job skill. For the first The job requirement weight of each skill. Contribution coefficient to capability gap The diagnostic terminal contributes a coefficient to the importance of the job; it can prioritize skills gaps based on their weights and develop improvement plans for skills that are essential for the job and have higher gap weights; the diagnostic terminal can simultaneously input structured knowledge constraints, student ability gap data and job requirement standards into a large model, and generate logic through standardized prompts. Combined with the student's existing ability base, it breaks down the learning into phased, feasible learning objectives, training plans and acceptance standards, and integrates them to obtain growth path planning suggestions.

[0067] The aforementioned competency gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement explicitly and structurally establishes the relationship between student training elements and job requirement elements by constructing an occupational competency knowledge graph. This provides a computable knowledge foundation for subsequent quantitative comparisons. Simultaneously, it maps student training information such as coursework, certification acquisition, and practical experience to the same job skill requirements dimension, eliminating the lack of structured semantic associations and quantitative mapping mechanisms between training elements and job requirements. This transforms competency diagnosis from subjective experience-based judgment to quantifiable vector comparison. Furthermore, it compares individual student competency vectors with job requirements dimension by dimension. The difference in mastery between competency requirement vectors accurately identifies specific competency gaps in students' various job skills and quantifies the degree of these gaps. Simultaneously, it calculates the overall matching degree, making the diagnostic results quantifiable and reproducible. Finally, by extracting structured knowledge related to competency gaps from a knowledge graph as a reliable context, it generates career path planning suggestions. The traceable entities and relationships in the knowledge graph form knowledge constraints on the generation process of the large model, ensuring that the generated content is verifiable and traceable. This solves the problems of output generalization and difficulty in accountability caused by the lack of structured knowledge constraints when existing large models directly generate career suggestions.

[0068] In one embodiment of the present invention, based on a professional competence knowledge graph, personal data and work skills of the target student are obtained. The degree of mastery of each work skill by the target student is determined according to the personal data, and the mastery levels are arranged into a vector form to obtain the student's individual competence vector, including:

[0069] S11. Based on the supporting relationship between courses and job skills in the professional competence knowledge graph, as well as the content of courses and the description of job skills, calculate the contribution coefficient of each course to each job skill.

[0070] Specifically, the contribution coefficient can be used to characterize the degree to which the knowledge content of a single course supports and covers the corresponding job skills. Unlike traditional course credit weights and course importance scores, the contribution coefficient can be generated through the semantic relationships of the professional ability knowledge graph.

[0071] For example, the diagnostic terminal can retrieve the fixed supporting associations of courses and job skills from the knowledge graph, filter out all course-skill pairings with direct or indirect supporting relationships, eliminate invalid pairings without any relationship, and reduce interference from invalid calculations; the diagnostic terminal can compare the degree of matching between the course content text and the corresponding job skill description text through a semantic similarity algorithm, and combine the course support priority and knowledge coverage dimension marked in the knowledge graph to quantify the contribution coefficient of each course to each related job skill.

[0072] S12. Obtain the grade data of the courses already taken by the target student, calculate the target student's mastery of each course based on the grade data, and weight and integrate the mastery level and contribution coefficient to obtain the target student's initial mastery of each job skill.

[0073] Specifically, the initial mastery of job skills can be calculated using students' course learning data and course contribution coefficients. The initial mastery of job skills reflects the level of students' mastery of job skills at the theoretical knowledge level, and can distinguish between theoretical basic ability and comprehensive practical ability. It is a core indicator for stratified quantification of students' skill level.

[0074] For example, the diagnostic terminal can obtain standardized grade data of all courses taken by the target student, perform normalization preprocessing on the raw grade data, and uniformly convert it into course mastery scores in the range of 0 to 1, eliminating calculation bias caused by differences in assessment scoring systems for different courses; the diagnostic terminal can calculate the initial skill mastery degree based on the contribution coefficient of each course to job skills using a weighted fusion algorithm; wherein, the calculation expression for the initial skill mastery degree can be:

[0075]

[0076] in, For the first Initial mastery of the job skill For students Normalized mastery scores for completed courses. For the first The course is for the first The contribution coefficient of each job skill To support the first The diagnostic terminal can combine the student's course mastery level with the course skill support weight to obtain the initial mastery results of each job skill.

[0077] S13. Obtain the certificate information and practical experience data of the target student. Based on the proof relationship between certificates and work skills, the manifestation relationship between practical experience data and work skills in the professional ability knowledge graph, and the certificate information and practical experience data, calculate the enhancement coefficient of certificates and practical experience on work skills. Combine the initial mastery and enhancement coefficient to obtain the final mastery of each work skill of the target student.

[0078] Specifically, the job skills enhancement coefficient can be used to characterize the effect of two types of practical training elements, namely professional certificates and practical experience, on the correction and improvement of students' mastery of theoretical skills. Unlike the course contribution coefficient, it focuses on the supplementary role of practical ability on theoretical skills. The job skills enhancement coefficient can be a parameter to achieve accurate correction of students' skill mastery and fit their actual performance capabilities.

[0079] For example, the diagnostic terminal can match the skill certification dimensions and certification levels corresponding to various certificates in the professional competence knowledge graph, as well as the skill practice scenarios and job performance depth corresponding to different types of practical experience. Combining the student's certificate acquisition level, practice duration, and practice job matching degree, it can quantify and generate certificate enhancement coefficient and practice experience enhancement coefficient. The diagnostic terminal can integrate the certificate enhancement coefficient and practice experience enhancement coefficient to obtain a comprehensive enhancement coefficient. The diagnostic terminal can also superimpose and integrate the comprehensive enhancement coefficient and the initial mastery of work skills to obtain the final mastery of each work skill.

[0080] S14. Arrange all job skills in the professional competence knowledge graph in a fixed order as a dimension index, and input the final mastery of each job skill of the target student into the position of the corresponding skill in the dimension index in sequence to generate the student's individual competence vector.

[0081] Specifically, the dimensional index can be a fixed sorted sequence built on all standardized work skills of the professional competence knowledge graph. The sorting rules are uniform and fixed and do not change with the individual student data. It provides a unified benchmark for the standardization and dimensional alignment of individual student competence vectors and is the basic system for realizing quantitative comparison of competence.

[0082] For example, the diagnostic terminal can match the final mastery of each student's work skills to the corresponding skill position in the dimension index one by one, and uniformly fill the default basic value for skill dimensions for which there is no corresponding mastery data. By matching and assigning values ​​through fixed dimension index, a standardized individual student ability vector is generated.

[0083] This application provides a capability gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement. Through progressive techniques such as hierarchical quantification, multi-source data fusion, and standardized dimensional mapping, it refines the calculation logic of students' work skill mastery layer by layer. First, it quantifies the supporting weights of theoretical skills based on course semantic matching. Then, it combines students' course grades to obtain the initial level of theoretical skill mastery. Further, it completes capability correction through the practical dimension coefficients of certificates and practical experience. Finally, it generates a standardized student capability vector based on a fixed and unified dimensional index. This method effectively overcomes the shortcomings of traditional student capability evaluation methods, such as chaotic dimensions, one-sided evaluation, strong subjectivity, and lack of quantifiable traceability. It achieves a comprehensive and accurate quantitative representation of students' theoretical and practical capabilities, ensuring the accuracy, objectivity, and reproducibility of subsequent job capability matching and gap diagnosis, and significantly improving the refinement and technical credibility of occupational capability diagnosis.

[0084] In one embodiment of the present invention, the initial mastery level and the enhancement coefficient are combined to obtain the calculation expression for the final mastery level of the target student for each job skill, which can be:

[0085] ;

[0086] in, For target students, For job skills, for right The ultimate degree of control, for right Initial level of mastery for The certificate information obtained for Data on practical experience To prove the relationship, To reflect the relationship, For certificate information, Data based on practical experience for right The enhancement coefficient, for right The enhancement coefficient.

[0087] For example, the diagnostic terminal can verify the correlation between students' certificates, practical experience and corresponding job skills based on the professional competence knowledge graph, and screen out effective practical enhancement coefficients. The diagnostic terminal can use the students' initial skill mastery as a theoretical benchmark, and carry out adaptive fusion correction by combining the enhancement coefficients of the two practical dimensions of certificates and practice. It can adjust the practical gain effect according to the skill differences of different theoretical mastery levels, strengthen the practical reinforcement effect for skills with weak theoretical knowledge, and constrain the gain range for skills with solid theoretical knowledge to avoid abnormal overflow of the ability quantification value. Through the dynamic fusion and calibration of theoretical and practical data, the final skill mastery that fits the students' real comprehensive performance ability can be obtained.

[0088] This application provides a capability gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement. Through an adaptive fusion of theoretical foundations and practical elements, it overcomes the shortcomings of traditional capability assessments, which involve simple superposition and homogeneous effects. It can dynamically adjust the skill enhancement gains corresponding to certificates and practical experience based on students' theoretical mastery of courses, effectively avoiding the problems of capability quantification distortion and numerical overflow. It achieves precise calibration and fusion of theoretical knowledge and practical ability, significantly improving the authenticity and relevance of students' final mastery of work skills. This provides a more reliable data foundation for subsequent construction of individual student capability vectors and precise diagnosis of job capability gaps.

[0089] In one embodiment of the present invention, such as Figure 2As shown, based on the individual student ability vector and the job ability requirement vector, the mastery level and demand intensity of the same skill dimension are compared. Job skills where the mastery level is lower than the demand intensity are considered as the ability gap of the target student. The overall matching degree between the target student and the target job is calculated. Integrating the ability gap and the overall matching degree, the ability gap diagnosis result is obtained, including:

[0090] S21. Align the student's individual ability vector and the job's ability requirement vector by dimension, and calculate the difference between the requirement intensity and the mastery level for each skill dimension to obtain the difference results for each skill dimension.

[0091] Specifically, vector dimension alignment can be used to unify the dimensionality of individual student ability vectors and job competency requirement vectors, avoiding comparison errors caused by disordered vector dimensions, missing dimensions, or inconsistent dimension order.

[0092] For example, the diagnostic terminal can perform dual-vector dimension alignment and single-dimensional difference calculation. Using the standardized skill dimension index as a unified benchmark, it performs dimension verification and adaptation alignment between the student's individual ability vector and the job ability requirement vector, eliminating the calculation deviation caused by vector dimension misalignment and dimension mismatch. After completing the dimension alignment, the diagnostic terminal can traverse each standardized job skill dimension, read the job skill requirement intensity value and the student skill mastery value corresponding to the current dimension one by one, and output the corresponding difference result for each dimension through the operation logic of subtracting the student mastery value from the job requirement intensity value.

[0093] S22. Arrange the difference results according to the order of the dimension indices to generate the difference vector.

[0094] Specifically, the difference vector can be generated by arranging the difference results of each skill dimension according to a unified and fixed rule, fully reflecting the matching differences between students' abilities and job requirements across all skill dimensions.

[0095] For example, the diagnostic terminal can fill the difference results into the corresponding dimension positions in sequence according to the globally unified skill dimension sorting rules, and the generated difference vector completely records the adaptation difference values ​​of all skill dimensions.

[0096] S23. Based on the difference vector, mark the job skills with a difference greater than zero on each skill dimension as competency gap skills, and take the difference corresponding to each competency gap skill as the competency gap value of the job skill. Summarize the competency gap skills and competency gap values ​​to obtain the competency gap set.

[0097] Specifically, the competency gap set can be used to collect skill data that students have shortcomings in the process of adapting to their jobs. The data is structured and summarized with skill dimensions and gap values ​​as core fields, and invalid dimension data that meet the adaptation standards or have excess capabilities are removed.

[0098] For example, the diagnostic terminal can complete the screening and summarization of competency gaps based on the standardized difference vector. It iterates through the difference values ​​of each dimension within the difference vector one by one, using a value greater than zero as the criterion for judging competency gaps. When the difference value of the corresponding skill dimension is greater than zero, it indicates that the student's skill mastery level in that dimension has not reached the job requirements. The diagnostic terminal can mark that skill dimension as a competency gap skill and define the difference value corresponding to the current dimension as the competency gap value of that job skill. After iterating through all skill dimensions, the diagnostic terminal can uniformly summarize all marked competency gap skills and their corresponding competency gap values, and remove the data of qualified skill dimensions with a difference less than or equal to zero to obtain the competency gap set.

[0099] S24. Based on the individual student ability vector and the job ability requirement vector, calculate the overall matching degree between the target student and the target job according to the numerical relationship between the mastery level and the demand intensity of each skill dimension.

[0100] Specifically, the overall job matching degree can be comprehensively represented by the ability adaptation of all skill dimensions, combined with the differences in job skill weights, to quantitatively represent the overall adaptability of students' overall abilities to the target job.

[0101] For example, the diagnostic terminal can comprehensively calculate the overall matching degree between the student and the target position based on the student's individual ability vector and the job ability requirement vector that are fully aligned in dimensions, combined with the job requirement weights of each skill dimension; the diagnostic terminal can calculate the overall matching degree through a weighted matching degree calculation method; wherein, the expression for calculating the overall matching degree can be:

[0102]

[0103] in, To assess the overall match between students and target positions, For the first The job requirement weight corresponding to each job skill For the first The numerical value of students' mastery of the job skills. For the first The job requirement intensity value for each skill. The total number of dimensions for standardized job skills.

[0104] S25. Based on the capability gap value, classify the severity level of each capability gap skill, and integrate the capability gap set, the severity level of the gap, and the overall matching degree into a structured capability gap diagnosis result.

[0105] Specifically, the severity level of the skill gap can be classified into levels based on the numerical value of the skill gap, which can distinguish the degree of difference between minor, moderate and severe skill gaps.

[0106] For example, the diagnostic terminal can perform standardized level classification based on the gap value corresponding to each skill gap, and classify all skill gaps into levels according to a preset numerical range threshold to distinguish the severity of different skill deficiencies. After completing the gap level classification, the diagnostic terminal can integrate the level of severity of the stratified skill gaps, the set of skill gaps including all skill gaps and their corresponding gap values, and the overall matching degree representing the global adaptation level in multiple dimensions to obtain a structured capability gap diagnosis result that combines details of individual shortcomings, differences in the stratification of shortcomings, and the global adaptation level.

[0107] This application provides a capability gap diagnosis method based on occupational knowledge graphs and large-scale model enhancement. Through a progressive processing logic of vector dimension alignment, difference quantification, gap screening, level stratification, and multi-dimensional result integration, it achieves standardized, refined, and structured quantitative evaluation of the capability gap between students and positions. It effectively solves the problems of chaotic capability comparison dimensions, ambiguous gap judgment, and lack of stratified evaluation in traditional methods. It can accurately distinguish between differences in single skill deficiencies and overall adaptability, significantly improving the structuring degree and reference value of capability gap diagnosis results. This provides stratified, accurate, and traceable diagnostic data support for subsequent personalized growth path planning.

[0108] In one embodiment of the present invention, based on the diagnostic results of the ability gap, structured knowledge related to the ability gap is obtained from the vocational ability knowledge graph. This structured knowledge is then input as a constraint into a large language model, and growth path planning suggestions for the ability gap are generated within the scope of the structured knowledge, including:

[0109] S31. Extract the feature information of each skill gap from the capability gap diagnosis results, and generate knowledge retrieval conditions based on the feature information; wherein, the knowledge retrieval conditions include a first retrieval condition with each skill gap as the retrieval subject and a second retrieval condition with the target position as the retrieval subject.

[0110] Specifically, multi-path knowledge retrieval conditions can be a two-dimensional retrieval constraint system. Multi-path knowledge retrieval conditions can adopt a dual-subject retrieval constraint logic of skill gaps and target positions, which can define the core knowledge range that matches students' shortcomings and job requirements.

[0111] For example, the diagnostic terminal can extract feature information such as skill name, gap level, and gap value corresponding to all skill gaps from the generated structured capability gap diagnostic results. It can then construct two-dimensional knowledge retrieval conditions by combining the skill feature information with the target job information. The first retrieval condition takes each skill gap as the retrieval subject and limits the retrieval scope to the knowledge system, training resources, and skill correlations corresponding to each skill gap. The second retrieval condition takes the target job as the retrieval subject and limits the retrieval scope to related knowledge such as job performance requirements, core skill standards, and job-matching training systems. The retrieval boundaries are defined by combining the two conditions.

[0112] S32. Using the first and second retrieval conditions in the knowledge retrieval criteria as inputs for multi-way queries, perform semantic retrieval and recall knowledge fragments that semantically match the knowledge retrieval criteria to obtain a recall result set.

[0113] Specifically, the recall result set can integrate two knowledge dimensions: skills enhancement and job matching, and replace traditional keyword matching with semantic association matching.

[0114] For example, the diagnostic terminal can perform semantic retrieval matching based on knowledge retrieval conditions and combined with a professional competence knowledge graph; to quantify the degree of knowledge matching and filter effective knowledge fragments, the diagnostic terminal can filter highly relevant knowledge through a semantic matching degree calculation function; wherein, the calculation expression of the semantic matching degree calculation function can be:

[0115]

[0116] in, To determine the semantic matching degree between the retrieved statement and the knowledge fragment in the graph, This is the semantic vector of the multi-way retrieval conditions. The semantic vector of the knowledge fragment to be matched in the professional competence knowledge graph; the diagnostic terminal can filter knowledge fragments with a semantic matching degree higher than a preset threshold, remove low-relevance and invalid redundant knowledge, and aggregate to obtain a standardized recall result set.

[0117] S33. Using the skills gaps and target positions as query objects, extract the entities and relationships associated with the skills gaps and the target positions from the professional competence knowledge graph to obtain the entity and relationship sequence, and transform the entity and relationship sequence into a structured knowledge context.

[0118] Specifically, structured knowledge contexts can be restructured around graph entities and relationships, with clear logical hierarchies and traceable origins, providing rigid knowledge constraints for generating content for large models.

[0119] For example, the diagnostic terminal can use the skills gap and the target position as dual query objects, traverse the occupational competency knowledge graph, and extract the courses, training, and certificate entities associated with the skills gap and their corresponding supporting relationships, as well as the job standards, performance requirements entities associated with the target position and their corresponding constraint relationships. The diagnostic terminal can summarize all valid entities and relationships, complete deduplication and normalization processing, and convert them into a structured knowledge context with a unified structure and logical order.

[0120] S34. Merge each recall result set and the structured knowledge context into a trusted knowledge context. Use the trusted knowledge context as the basis for generation and input it into the large language model to generate growth path planning suggestions within the scope of the trusted knowledge context.

[0121] Specifically, credible knowledge context can integrate semantic recall results with graph-structured knowledge to fundamentally solve the problems of unfounded output and poor adaptability of large models; large language models can be pre-trained AI models with deep semantic understanding, knowledge inductive reasoning and structured text generation capabilities. They can complete adaptive and personalized content generation under the constraints of external credible structured knowledge, and can make logical deductions based on students' ability shortcomings and job requirements, effectively avoiding the defects of templated output and excessive generalization.

[0122] For example, the diagnostic terminal can integrate the retrieved result set and the structured knowledge context, eliminate duplicate content, fill in missing related information, and construct a complete and reliable knowledge context. The diagnostic terminal can use the reliable knowledge context as the sole constraint input to the large language model, limiting the large language model to logical deduction and content generation only within the scope of reliable knowledge. The diagnostic terminal can also use the large language model to combine the severity of students' skill gaps with job weights to generate targeted, personalized growth path planning suggestions that are tailored to students' learning situations, adaptable to job requirements, and feasible.

[0123] This application provides a capability gap diagnosis method based on professional knowledge graphs and large-scale model enhancement. Through a progressive logic of dual-path precise retrieval, semantic quantitative screening, graph structured parsing, and credible knowledge constraint generation, it constructs a dedicated architecture that rigidly constrains the output of the large-scale model using knowledge graphs. It ensures the accuracy of knowledge retrieval through a quantitative semantic matching mechanism and limits the model generation boundary by relying on structured credible knowledge. This effectively solves the defects of traditional large-scale model career planning content generalization, lack of basis, and weak adaptability, and significantly improves the credibility, accuracy, and practicality of growth path planning suggestions.

[0124] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0125] Based on the same inventive concept, this application also provides a capability gap diagnosis system based on occupational knowledge graphs and large model enhancement for implementing the capability gap diagnosis method based on occupational knowledge graphs and large model enhancement described above. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the capability gap diagnosis system based on occupational knowledge graphs and large model enhancement provided below can be found in the limitations of the capability gap diagnosis method based on occupational knowledge graphs and large model enhancement described above, and will not be repeated here.

[0126] In one exemplary embodiment, such as Figure 3 As shown, a capability gap diagnosis system 400 based on occupational knowledge graphs and large model enhancement is provided to implement the methods in the above-described method embodiments, including:

[0127] The knowledge graph construction module 401 is used to obtain the elements of student ability development, the elements of job ability requirements, and the relationship between the elements of student ability development and the elements of job ability requirements, and to construct a professional ability knowledge graph based on the elements of student ability development, the elements of job ability requirements, and the relationship.

[0128] The student competency modeling module 402 is used to acquire the target student's personal data and work skills based on the professional competency knowledge graph, determine the target student's mastery of each work skill based on the personal data, arrange the mastery level into a vector form, and obtain the student's individual competency vector.

[0129] The job requirement modeling module 403 is used to obtain the target job specified by the user, calculate the demand intensity of the job skills for the target job based on the correlation information between the target job and the job skills, and arrange the demand intensity into a vector form to obtain the job capability requirement vector.

[0130] The competency gap diagnosis module 404 is used to compare the mastery level and demand intensity of the same skill dimension based on the individual student's competency vector and the job competency demand vector. The module identifies the job skills with a mastery level lower than the demand intensity as the competency gap of the target student, calculates the overall matching degree between the target student and the target job, integrates the competency gap and the overall matching degree, and obtains the competency gap diagnosis result.

[0131] The growth path planning module 405 is used to extract structured knowledge related to the ability gap from the professional ability knowledge graph based on the ability gap diagnosis results, input the structured knowledge as a constraint into the large language model, and generate growth path planning suggestions for the ability gap within the scope of the structured knowledge.

[0132] In one embodiment of the present invention, the student ability modeling module 402 can also be used for:

[0133] S11. Based on the supporting relationship between courses and job skills in the professional competence knowledge graph, as well as the content of courses and the description of job skills, calculate the contribution coefficient of each course to each job skill.

[0134] S12. Obtain the grade data of the courses already taken by the target student, calculate the target student's mastery of each course based on the grade data, and weight and integrate the mastery level and contribution coefficient to obtain the target student's initial mastery of each job skill.

[0135] S13. Obtain the certificate information and practical experience data of the target student. Based on the proof relationship between certificates and work skills in the professional ability knowledge graph and the manifestation relationship between practical experience and work skills, as well as the certificate information and practical experience data, calculate the enhancement coefficient of certificates and practical experience on work skills. Combine the initial mastery and enhancement coefficient to obtain the final mastery of each work skill of the target student.

[0136] S14. Arrange all job skills in the professional competence knowledge graph in a fixed order as a dimension index, and input the final mastery of each job skill of the target student into the position of the corresponding skill in the dimension index in sequence to generate the student's individual competence vector.

[0137] In one embodiment of the present invention, the initial mastery level and the enhancement coefficient are combined to obtain the calculation expression for the final mastery level of the target student for each job skill:

[0138] ;

[0139] in, For target students, For job skills, for right The ultimate degree of control, for right Initial level of mastery for The certificate information obtained for Data on practical experience To prove the relationship, To reflect the relationship, for right The enhancement coefficient, for right The enhancement coefficient.

[0140] In one embodiment of the present invention, the capability gap diagnosis module 404 can also be used for:

[0141] S21. Align the student's individual ability vector and the job ability requirement vector by dimension, and calculate the difference between the requirement intensity and the mastery level for each skill dimension to obtain the difference results for each skill dimension.

[0142] S22. Arrange the difference results according to the dimension index to generate a difference vector;

[0143] S23. Based on the difference vector, mark the work skills with a difference greater than zero in each skill dimension as competency gap skills, and take the difference corresponding to each competency gap skill as the competency gap value of the work skill. Summarize the competency gap skills and competency gap values ​​to obtain the competency gap set.

[0144] S24. Based on the individual student ability vector and the job ability requirement vector, calculate the overall matching degree between the target student and the target job according to the numerical relationship between the mastery level and the demand intensity of each skill dimension.

[0145] S25. Based on the capability gap value, classify the severity level of each capability gap skill, and integrate the capability gap set, the severity level of the gap, and the overall matching degree into a structured capability gap diagnosis result.

[0146] In one embodiment of the present invention, the growth path planning module 405 can also be used for:

[0147] S31. Extract the feature information of each skill gap from the capability gap diagnosis results, and generate knowledge retrieval conditions based on the feature information; wherein, the knowledge retrieval conditions include a first retrieval condition with each skill gap as the retrieval subject and a second retrieval condition with the target position as the retrieval subject;

[0148] S32. Using the first and second retrieval conditions in the knowledge retrieval conditions as inputs for multi-way queries, semantic retrieval is performed and knowledge fragments that semantically match the knowledge retrieval conditions are recalled to obtain a recall result set.

[0149] S33. Using the skills gaps and target positions as query objects, extract the entities and relationships associated with the skills gaps and the target positions from the professional competence knowledge graph to obtain the entity and relationship sequence, and transform the entity and relationship sequence into a structured knowledge context.

[0150] S34. Merge each recall result set and structured knowledge context into a trusted knowledge context. Use the trusted knowledge context as the basis for generation and input it into the large language model to generate growth path planning suggestions within the scope of the trusted knowledge context.

[0151] In one embodiment, this application provides a computer device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0152] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0153] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0154] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A method for diagnosing competency gaps based on occupational knowledge graphs and large-scale model enhancement, characterized in that, The method includes: S1. Obtain student ability development elements, job ability requirement elements, and the relationship between the student ability development elements and the job ability requirement elements, and construct a professional ability knowledge graph based on the student ability development elements, the job ability requirement elements, and the relationship. S2. Based on the professional ability knowledge graph, obtain the target student's personal data and work skills, determine the target student's mastery level in each of the work skills according to the personal data, and arrange the mastery level into a vector form to obtain the student's individual ability vector. S3. Obtain the target job specified by the user, calculate the demand intensity of the job skills for the target job based on the correlation information between the target job and the job skills, and arrange the demand intensity into a vector form to obtain the job competency demand vector. S4. Based on the student's individual ability vector and the job's ability requirement vector, compare the mastery level and the demand intensity on the same skill dimension, take the job skills whose mastery level is lower than the demand intensity as the ability gap of the target student, calculate the overall matching degree between the target student and the target job, integrate the ability gap and the overall matching degree to obtain the ability gap diagnosis result. S5. Based on the diagnostic results of the ability gap, obtain structured knowledge related to the ability gap from the professional ability knowledge graph, input the structured knowledge as a constraint into the large language model, and generate growth path planning suggestions for the ability gap within the scope of the structured knowledge.

2. The method according to claim 1, characterized in that, Based on the professional competence knowledge graph, the individual data and work skills of the target student are obtained. The degree of mastery of each work skill by the target student is determined according to the individual data. The mastery degrees are then arranged into a vector form to obtain the student's individual competence vector, including: S11. Based on the supporting relationship between courses and job skills in the professional competence knowledge graph, as well as the content of the courses and the description of the job skills, calculate the contribution coefficient of each course to each job skill. S12. Obtain the grade data of the courses already taken by the target student, calculate the target student's mastery of each course based on the grade data, and perform a weighted fusion of the mastery level and the contribution coefficient to obtain the target student's initial mastery of each of the work skills. S13. Obtain the certificate information and practical experience data already obtained by the target student. Based on the proof relationship between certificates and work skills in the professional ability knowledge graph, the manifestation relationship between practical experience data and work skills, and the certificate information and practical experience data, calculate the enhancement coefficient of the certificates and practical experience on work skills. Combine the initial mastery and the enhancement coefficient to obtain the final mastery of each work skill by the target student. S14. Arrange all the job skills in the professional ability knowledge graph in a fixed order as a dimension index, and input the final mastery degree of the target student for each job skill into the position of the corresponding skill in the dimension index in sequence to generate the student's individual ability vector.

3. The method according to claim 2, characterized in that, The expression for calculating the final mastery of each job skill by fusing the initial mastery level and the enhancement coefficient is as follows: ； in, For the target students, For the aforementioned job skills, for right The final degree of mastery mentioned above, For the right The initial level of mastery, for The obtained certificate information, for The data on the aforementioned practical experience, For the proof relation, To embody the relationship, For the certificate information, For the data of the aforementioned practical experience, for right The enhancement coefficient, for right The enhancement coefficient.

4. The method according to claim 1, characterized in that, The process involves comparing the level of mastery and the intensity of demand for the same skill dimension based on the student's individual ability vector and the job's ability requirement vector. Skills where the level of mastery is lower than the intensity of demand are identified as the ability gap for the target student. The overall matching degree between the target student and the target job is calculated. The ability gap and the overall matching degree are then integrated to obtain an ability gap diagnosis result, including: S21. Align the student individual ability vector and the job ability requirement vector by dimension, and calculate the difference between the requirement intensity and the mastery level for each skill dimension to obtain the difference results for each skill dimension. S22. Arrange the difference results according to the order of the dimension indices to generate a difference vector; S23. Based on the difference vector, mark the work skills with differences greater than zero on each of the skill dimensions as competency gap skills, and use the difference corresponding to each competency gap skill as the competency gap value of the work skill. Summarize the competency gap skills and the competency gap values ​​to obtain a competency gap set. S24. Based on the student's individual ability vector and the job's ability requirement vector, calculate the overall matching degree between the target student and the target job according to the numerical relationship between the mastery level and the demand intensity in each of the skill dimensions; S25. Based on the capability gap value, classify the severity level of each capability gap skill, and integrate the capability gap set, the severity level of the gap, and the overall matching degree into a structured capability gap diagnosis result.

5. The method according to claim 1, characterized in that, Based on the diagnostic results of the ability gap, structured knowledge related to the ability gap is obtained from the vocational ability knowledge graph. This structured knowledge is then input into a large language model as a constraint. Within the scope of this structured knowledge, growth path planning suggestions for the ability gap are generated, including: S31. Extract feature information of each skill gap from the capability gap diagnosis results, and generate knowledge retrieval conditions based on the feature information; wherein, the knowledge retrieval conditions include a first retrieval condition with each skill gap as the retrieval subject and a second retrieval condition with the target position as the retrieval subject; S32. Using the first retrieval condition and the second retrieval condition in the knowledge retrieval conditions as multi-way query inputs, perform semantic retrieval and recall knowledge fragments that semantically match the knowledge retrieval conditions to obtain a recall result set; S33. Using the skills with skill gaps and the target job as query objects, extract entities and relationships associated with the skills with skill gaps and entities and relationships associated with the target job from the professional competence knowledge graph to obtain a sequence of entities and relationships, and transform the sequence of entities and relationships into a structured knowledge context. S34. Merge each of the recall result sets and the structured knowledge context into a trusted knowledge context, use the trusted knowledge context as the basis for generation and input it into the large language model, and generate the growth path planning suggestion within the scope of the trusted knowledge context.

6. A capability gap diagnosis system based on occupational knowledge graphs and large-scale model enhancement, used to implement the method of any one of claims 1 to 5, characterized in that, The system includes: The knowledge graph construction module is used to obtain student ability development elements, job ability requirement elements, and the relationship between the student ability development elements and the job ability requirement elements, and to construct a professional ability knowledge graph based on the student ability development elements, the job ability requirement elements, and the relationship. The student ability modeling module is used to obtain the target student's personal data and work skills based on the occupational ability knowledge graph, determine the target student's mastery level in each of the work skills based on the personal data, and arrange the mastery level into a vector form to obtain the student's individual ability vector. The job requirement modeling module is used to obtain the target job specified by the user, calculate the demand intensity of the job skills for the target job based on the correlation information between the target job and the job skills, and arrange the demand intensity into a vector form to obtain the job capability requirement vector. The competency gap diagnosis module is used to compare the mastery level and demand intensity of the same skill dimension based on the individual student competency vector and the job competency demand vector, identify the job skills where the mastery level is lower than the demand intensity as the competency gap of the target student, calculate the overall matching degree between the target student and the target job, integrate the competency gap and the overall matching degree to obtain the competency gap diagnosis result. The growth path planning module is used to obtain structured knowledge related to the ability gap from the professional ability knowledge graph based on the ability gap diagnosis results, input the structured knowledge as a constraint into the large language model, and generate growth path planning suggestions for the ability gap within the scope of the structured knowledge.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5.

Images