Vocational ability optimization scheme recommendation method and device, computer equipment and medium

Abstract

The invention relates to the technical field of performance optimization, and discloses a vocational ability optimization scheme recommendation method and device, computer equipment and a medium, and the method comprises the steps: obtaining vocational ability to-be-evaluated data of a to-be-evaluated object; performing vocational ability assessment on the to-be-assessed object based on the vocational ability to-be-assessed data and a preset vocational ability assessment model to obtain a vocational ability assessment result; constructing a vocational ability optimization recommendation scheme knowledge graph based on a vocational ability assessment result and a to-be-promoted direction selection result after self-assessment of the to-be-assessed object; and obtaining a recommended vocational ability optimization scheme based on the vocational ability optimization recommendation scheme knowledge graph. According to the invention, the work ability and level of the counselor can be improved in a targeted manner, and better service and support are provided for the growth and development of students.

Description

Technical Field

[0001] The present invention relates to the field of performance optimization technology, and in particular to a method, device, computer equipment and medium for recommending a professional capability optimization plan. Background Art

[0002] With the widespread adoption and deepening of digital education platforms, higher vocational colleges are playing an increasingly important role in cultivating students' professional skills and overall well-being. As the core of student management at higher vocational colleges, counselors' professional competence directly impacts students' growth and development. However, the work of higher vocational college counselors is complex, encompassing daily student management, mental health education, career planning guidance, and other aspects. Efficiently and scientifically improving the professional competence of higher vocational college counselors has become a major challenge facing higher vocational college management.

[0003] Traditionally, counselors' professional development has primarily relied on experience transfer, regular training, and self-study. While these approaches can improve counselors' overall quality to a certain extent, they lack specificity and personalization, making it difficult to meet the specific needs of different counselors in their work. Furthermore, the vast amount of information available, including work data and student feedback from higher vocational counselors, is often underutilized, leading to wasted resources and information silos. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, computer equipment and medium for recommending a career optimization plan to solve the problem that the existing technology lacks pertinence in the recommendation of career optimization for higher vocational counselors.

[0005] In a first aspect, the present invention provides a method for recommending a professional capability optimization plan, the method comprising:

[0006] Obtain the professional competence assessment data of the subject to be assessed;

[0007] Based on the occupational ability assessment data and the preset occupational ability assessment model, the occupational ability assessment of the assessment object is performed to obtain the occupational ability assessment results;

[0008] Construct a knowledge graph of recommended solutions for optimizing professional competence based on the professional competence assessment results and the results of the direction of improvement selected by the assessed subjects after self-assessment;

[0009] The recommended career capability optimization plan is obtained based on the career capability optimization recommendation plan knowledge graph.

[0010] The present invention provides a method for recommending a career optimization plan, which obtains the career ability data to be evaluated by the higher vocational counselors, and provides a basis for accurately learning the relationship between the career ability performance characteristics of the higher vocational counselors and the career ability evaluation results based on the comprehensive career ability performance data of the higher vocational counselors, so as to accurately predict the career ability level of the higher vocational counselors; based on the career ability to be evaluated data and a preset career ability evaluation model, the career ability of the subject to be evaluated is evaluated, and the career ability evaluation result is obtained; based on the career ability evaluation result and the result of the direction to be promoted after the self-evaluation of the subject to be evaluated, a knowledge map of the career optimization recommendation plan is constructed; the result of the direction to be promoted obtained after the self-evaluation of the subject to be evaluated is obtained, so as to obtain the career to be self-selected and optimized by the higher vocational counselors. Ability direction provides a foundation for higher vocational counselors to actively optimize their own professional abilities based on subjective initiative. The knowledge graph of professional ability optimization recommendation plans, comprehensive professional ability assessment results and the results of selection of directions to be improved, are matched using the knowledge graph of professional ability optimization recommendation plans for higher vocational counselors to obtain the corresponding recommended professional ability optimization plans for higher vocational counselors; the recommended professional ability optimization plans based on the knowledge graph of professional ability optimization recommendation plans can accurately assess the professional ability level of higher vocational counselors and recommend their professional ability optimization plans in a targeted manner, so as to effectively improve the work ability and level of counselors, provide better services and support for students' growth and development, and solve the problem of lack of targeted recommendations for professional ability optimization of higher vocational counselors in existing technologies.

[0011] In an optional embodiment, the professional competence data to be assessed includes quantitative work performance data, quantitative student satisfaction data, and quantitative training participation data. Before performing a professional competence assessment on the subject to be assessed based on the professional competence data to be assessed and a preset professional competence assessment model, the method further includes:

[0012] The quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation were subjected to z-score standardization respectively, and the quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation after z-score standardization were used as the data to be evaluated for professional ability.

[0013] The present invention provides a method for recommending professional competency optimization plans. Through z-score standardization, the method converts quantitative data on work performance, student satisfaction, and training participation into standard normally distributed data with a mean of 0, using standard deviation as the unit. This eliminates interference caused by dimensional differences, provides an equal basis for comparison of different types of data, ensures that each data type plays a reasonable role in professional competency assessment, and avoids bias in assessment results caused by dimensionality issues. The standardized data distribution is more stable, reducing the impact of outliers in the original data.

[0014] In an optional embodiment, a professional competence assessment is performed on the subject to be assessed based on the professional competence data to be assessed and a preset professional competence assessment model to obtain a professional competence assessment result, including:

[0015] The standardized quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation are input into the preset professional ability assessment model respectively to obtain the quantitative scores of work performance, quantitative scores of student satisfaction and quantitative scores of training participation;

[0016] The three-dimensional scores of work performance quantitative scores, student satisfaction quantitative scores and training participation quantitative scores are used as the results of professional ability assessment.

[0017] The present invention provides a method for recommending a career capability optimization plan. The method inputs data from three different dimensions, namely work performance, student satisfaction, and training participation, into a preset model. The method evaluates the career capability of higher vocational counselors from multiple aspects, such as work results, feedback from service recipients, and self-improvement enthusiasm, thus avoiding the one-sidedness of single indicator evaluation. By comprehensively considering the quantitative scores of different dimensions, the career capability status of higher vocational counselors can be more comprehensively and accurately portrayed, providing more targeted guidance for subsequent career development. Using three-dimensional scores as the result of career capability assessment, the complex career capability status can be presented in an intuitive and concise manner.

[0018] In an optional embodiment, a knowledge graph of a career capability optimization recommendation solution is constructed based on the career capability assessment results and the results of the assessment subject's self-assessment of the direction to be improved, including:

[0019] Based on the professional competency assessment results, the results of the direction of improvement selected by the subject after self-assessment, and the preset personalized professional competency optimization plan, a structured professional competency optimization plan data source and an unstructured professional competency optimization plan data source are obtained;

[0020] The structured occupational capability optimization program data source and the unstructured occupational capability optimization program data source are cleaned separately to remove duplicate, erroneous and inconsistent data, thereby obtaining the standby structured occupational capability optimization program data source and the standby unstructured occupational capability optimization program data source;

[0021] Based on the preset management archive database and the preset professional capability optimization solution database, the data source of the structured professional capability optimization solution to be used is aligned with entities and filled with attributes to obtain knowledge fusion data;

[0022] Perform entity extraction, relationship extraction and attribute extraction on the unstructured professional capability optimization solution data source to obtain knowledge extraction data;

[0023] After performing coreference resolution and entity disambiguation on the knowledge fusion data and knowledge extraction data, the relationships between entities are obtained, and an initial database is constructed based on the relationships between entities.

[0024] The entities and relations of the initial database are converted into entity triples in the database, and a knowledge graph of career capability optimization recommendation solutions is constructed based on the entity triples.

[0025] The present invention provides a method for recommending a career capability optimization plan, which integrates the career capability assessment results, the selection results of the direction to be improved after the self-assessment of the subject to be assessed, and the structured and unstructured data sources of the preset personalized career capability optimization plan, breaks down data barriers, and integrates multi-channel information. Data cleaning is performed on the structured and unstructured data sources respectively to remove duplicate, erroneous and inconsistent data. This effectively avoids errors or inaccuracies in the knowledge graph caused by data quality problems, ensures that the knowledge graph constructed subsequently is based on high-quality data, performs entity alignment and attribute filling on the structured data source to be used, and extracts entities, relationships and attributes from the unstructured data source to be used, and converts scattered data into structured knowledge. Through knowledge fusion and extraction, the potential correlation relationship between the data is excavated, so that the knowledge graph can clearly present the logical connection between the career capability assessment results, the direction to be improved and the optimization plan, performs co-reference elimination and entity disambiguation on the knowledge fusion data and the knowledge extraction data, solves the problem of multiple expressions or ambiguity of the same entity, and ensures the uniqueness and accuracy of the entities and relationships in the knowledge graph. The initial database's entities and relationships are converted into triples of each entity within the database to construct a knowledge graph. This structured storage facilitates knowledge querying, updating, and maintenance. Furthermore, the triple-based knowledge graph structure is easily integrated with other systems, enabling rapid and efficient recommendation of appropriate optimization solutions based on professional competency assessment results and areas for improvement.

[0026] In an optional implementation, entity extraction, relationship extraction, and attribute extraction are performed on the unstructured professional capability optimization solution data source to obtain knowledge extraction data, including:

[0027] Identify and locate entities in the data source of the unstructured professional competence optimization plan to be used, and classify the entities into predefined entity categories as entity extraction data. The predefined entity categories include personalized professional competence optimization plans, professional competence assessment results, and results of selection of areas for improvement.

[0028] Based on statistical learning technology, deep learning algorithms are used to learn the characteristics of the relationship between entities, and the learned characteristics are used to extract the relationship between entities from the unstructured professional ability optimization solution data source to be used as relationship extraction data;

[0029] Utilizing the attribute extraction deep learning network, the attributes of each entity are extracted from the unstructured occupational capability optimization solution data source as attribute extraction data;

[0030] Entity extraction data, relationship extraction data and attribute extraction data are used as knowledge extraction data.

[0031] The present invention provides a method for recommending a career capability optimization plan, which can quickly sort out the key information in the data by identifying and locating entities in unstructured data sources and dividing them into predefined entity categories, such as personalized career capability optimization plans, career capability assessment results, and results of selecting directions for improvement. This targeted entity classification method lays a clear framework foundation for the subsequent construction of a knowledge graph. It comprehensively uses statistical learning techniques and deep learning algorithms to extract the relationships between entities from unstructured data, so that the knowledge graph can more comprehensively and accurately reflect the logical connections between the various elements in the career capability optimization process, providing strong support for accurately recommending optimization plans. Entity attributes are extracted using an attribute extraction deep learning network. The network has powerful feature learning capabilities and can accurately extract detailed attribute information of each entity from complex unstructured texts. Integrating entity extraction, relationship extraction, and attribute extraction into a complete data processing process can achieve efficient conversion from unstructured data sources to knowledge extraction data, which can greatly enhance the practicality of the knowledge graph for recommending career capability optimization plans.

[0032] In an optional embodiment, the recommended professional capability optimization solution is obtained based on the professional capability optimization recommendation solution knowledge graph, including:

[0033] Obtaining a set of objects to be evaluated, setting a bipartite graph between each object to be evaluated in the set of objects to be evaluated and the professional capability optimization solution, and obtaining an interactive relationship between the object to be evaluated and the professional capability optimization solution based on the bipartite graph;

[0034] Add the set of objects to be evaluated and their interaction relationships to the knowledge graph of the professional ability optimization recommendation plan to form an attention network of the professional ability optimization knowledge graph; the attention network of the professional ability optimization knowledge graph includes nodes and edges;

[0035] The preset network knowledge representation model is used to learn the vector representations of nodes and edges in the attention network of the professional ability optimization knowledge graph, and the first-order embedding vector representations of the object to be evaluated and the professional ability optimization solution are obtained;

[0036] Based on the first-order embedding vector representations of the objects to be evaluated and the professional ability optimization solutions, a preset attention embedding propagation module is used to update the node representations in the attention network of the professional ability optimization knowledge graph, generating aggregated attention vector representations of multiple objects to be evaluated and professional ability optimization solutions. The preset attention embedding propagation module includes several propagation layers, each of which recursively propagates the embedding vectors of neighboring nodes to update the node representation;

[0037] The first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan, the node representation of each propagation layer, and the aggregated attention vector representation are spliced to obtain the matching score of each object to be evaluated and the professional ability optimization plan;

[0038] The occupational capability optimization plan corresponding to the object to be evaluated when the matching score is the highest will be used as the recommended occupational capability optimization plan.

[0039] The present invention provides a method for recommending a professional capability optimization plan, which presents complex corresponding relationships in an intuitive and clear graphical structure by constructing a bipartite graph between the object to be evaluated and the professional capability optimization plan. This method can quickly locate the relationship between the object to be evaluated and the optimization plan, form an attention network of the professional capability optimization knowledge graph, and utilize the characteristics of the attention mechanism to automatically focus on important nodes and edges in the knowledge graph that are related to the recommendation of professional capability optimization plans. The preset network knowledge representation model is used to learn the vector representation of nodes and edges, and the object to be evaluated and the professional capability optimization plan are converted into a low-dimensional dense vector form, which can effectively mine the potential features and semantic information behind the data. These vector representations not only contain the explicit attributes of the entity, but also capture the complex implicit relationships between entities. The preset attention embedding propagation module updates the node representation by recursively propagating the embedding vectors of neighboring nodes through multiple layers. The propagation of each layer is a further fusion and refinement of knowledge. This layered propagation mechanism aggregates information at varying depths and breadths, allowing node representations to continuously absorb knowledge from surrounding related nodes, thereby more comprehensively reflecting the combined characteristics of the evaluated object and the optimization solution. It comprehensively considers the first-order embedding vector representation, the node representations of each propagation layer, and the aggregated attention vector representation to calculate the matching score from multiple dimensions. This multi-dimensional evaluation method comprehensively measures the fit between the evaluated object and the professional competency optimization solution, avoiding the one-sidedness of single-dimensional evaluation. Ultimately, the solution with the highest matching score is recommended, significantly improving the adaptability of the recommended solution to the actual needs of the evaluated object and achieving precise recommendations.

[0040] In an optional embodiment, based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization solution, a preset attention embedding propagation module is used to update the node representation in the attention network of the professional ability optimization knowledge graph to generate an aggregated attention vector representation of multiple objects to be evaluated and professional ability optimization solutions, including:

[0041] Construct a set of head entity triples;

[0042] Based on the head entity triple set, the information content of the tail entity vector of the head entity triple is obtained;

[0043] Based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan and the information content of the tail entity vector of the head entity triplet, the representations of the object to be evaluated and the professional ability optimization plan output by all propagation layers are aggregated respectively to obtain the aggregated attention vector representation of the object to be evaluated and the professional ability optimization plan.

[0044] The present invention provides a method for recommending a professional capability optimization solution, which presents the entity relationships in the professional capability optimization knowledge graph attention network in the form of structured triples by constructing a head entity triple set. This method can systematically integrate the object to be evaluated, the professional capability optimization solution and the associated information between them, so that the scattered knowledge forms an orderly logical structure, calculates the information content of the tail entity vector based on the head entity triple set, and measures the importance of each tail entity in the triple relationship in a quantitative manner. In this way, key information closely related to the object to be evaluated and the professional capability optimization solution can be effectively identified, avoiding interference from irrelevant or secondary information during the information aggregation process. Taking into account the first-order embedding vector representation of the object to be evaluated and the professional capability optimization solution, as well as the information content of the tail entity vector of the head entity triple, the representations output by all propagation layers are aggregated. This multi-factor aggregation method not only retains the potential feature information of the entity itself (represented by the first-order embedding vector), but also incorporates the importance information of the relationship between entities (through the information content of the tail entity vector). The fusion of multi-dimensional information enables the aggregated attention vector representation to more comprehensively and accurately reflect the comprehensive characteristics of the object to be evaluated and the professional competency optimization plan. Using a pre-set attention embedding propagation module, hierarchical aggregation is performed, with each layer updating and propagating information based on the previous layer, gradually integrating the knowledge of neighboring nodes. This hierarchical aggregation mechanism can mine the associations between entities at different depths, allowing node representations to continuously absorb information from surrounding related nodes, achieving effective knowledge dissemination and deep integration within the network. The resulting aggregated attention vector representation of the object to be evaluated and the professional competency optimization plan more accurately depicts the relationship and characteristics between them.

[0045] In a second aspect, the present invention provides a device for recommending a career capability optimization plan, the device comprising:

[0046] The module for obtaining data to be evaluated is used to obtain the data on the professional ability to be evaluated of the object to be evaluated;

[0047] A professional competency assessment module is used to assess the professional competency of the subject to be assessed based on the professional competency data to be assessed and a preset professional competency assessment model, and obtain professional competency assessment results;

[0048] A knowledge graph construction module is used to construct a knowledge graph of career capability optimization recommendation solutions based on the career capability assessment results and the results of the assessment subjects' self-assessment of the direction of improvement;

[0049] The professional ability optimization plan recommendation module is used to obtain recommended professional ability optimization plans based on the professional ability optimization recommendation plan knowledge graph.

[0050] In a third aspect, the present invention provides a computer device comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to execute the method for recommending a professional capability optimization plan according to the first aspect or any corresponding embodiment thereof.

[0051] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the method for recommending a professional competence optimization plan according to the first aspect or any corresponding embodiment thereof.

[0052] In a fifth aspect, the present invention provides a computer program product comprising computer instructions for enabling a computer to execute the method for recommending a professional competence optimization plan according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 1 is a flow chart of a method for recommending a career capability optimization solution according to an embodiment of the present invention;

[0055] Figure 2 is a flowchart of another method for recommending a professional capability optimization solution according to an embodiment of the present invention;

[0056] Figure 3 is a flowchart of another method for recommending a professional capability optimization solution according to an embodiment of the present invention;

[0057] Figure 4is a flowchart of another method for recommending a professional capability optimization solution according to an embodiment of the present invention;

[0058] Figure 5 is a structural block diagram of a device for recommending a career capability optimization solution according to an embodiment of the present invention;

[0059] Figure 6 Schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0060] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0061] According to an embodiment of the present invention, an embodiment of a method for recommending a professional competence optimization plan is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0062] In this embodiment, a method for recommending a career capability optimization plan is provided, which can be used in the above-mentioned computer device. Figure 1 is a flow chart of a method for recommending a career capability optimization solution according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0063] Step S101: Obtain the professional ability assessment data of the subject to be assessed.

[0064] Specifically, the subjects to be evaluated may be higher vocational counselors and higher vocational administrative staff, etc. The professional competence data to be evaluated may include quantitative data on work performance, quantitative data on student satisfaction, and quantitative data on training participation.

[0065] Quantitative data on work performance include the effectiveness of student academic guidance, academic style construction and student affairs management; quantitative data on student satisfaction include student satisfaction assessment scores and student evaluation ratios; quantitative data on training participation include the number of training participants, training participation rate, number and quality of training courses.

[0066] In this embodiment, student academic guidance is reflected in one-on-one tutoring, study groups, etc. to help students clarify their learning goals and improve their learning motivation, thereby reducing the failure rate. The specific indicator of the effectiveness of student academic guidance is set as the percentage reduction in the failure rate of the responsible students compared with the previous semester; the construction of academic style is reflected through the students' learning attitude and learning atmosphere, and the specific indicators of the effectiveness of academic style construction are set as student attendance rate, class participation and homework submission rate; student affairs management includes class activity organization and student discipline and violation handling, and the specific indicators are class activity participation rate and the reduction in student discipline and violation incidents; the student satisfaction assessment score can directly reflect students' high The degree of recognition of the work of vocational counselors; the proportion of student evaluation is also an important aspect of measuring student satisfaction in the assessment of higher vocational counselors; the number of training participants includes the actual number of online and offline participants, which can effectively reflect the scale and influence of the training; the training participation rate refers to the ratio of the number of people who actually participate in the training to the total number of people who should participate in the training, which can effectively reflect the enthusiasm of the trainees to participate in the training; the number of training courses can effectively measure the training workload organized by higher vocational counselors. The quality of training courses is evaluated and scored from three aspects: teaching content, teaching methods and teaching resources, which can evaluate the workload, work efficiency and practicality, accuracy and depth of the training content of higher vocational counselors.

[0067] Step S102 : performing a professional competence assessment on the subject to be assessed based on the professional competence assessment data and a preset professional competence assessment model to obtain a professional competence assessment result.

[0068] Specifically, the preset professional competency assessment model is a pre-trained professional competency assessment model, such as a support vector machine (SVM) model. The support vector machine model (SVM model for short) is a supervised learning algorithm widely used in machine learning, primarily for classification and regression analysis. The core idea of the SVM is to separate data points of different categories as closely as possible by finding an optimal hyperplane in the feature space that maximizes the margin, i.e., the interval, between sample points of different categories.

[0069] Taking higher vocational counselors as an example, the quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation of higher vocational counselors during their work period are input into the trained SVM model. The model can accurately identify the quantitative scores of work performance, quantitative scores of student satisfaction and quantitative scores of training participation of higher vocational counselors corresponding to the input professional ability evaluation data, and use the corresponding quantitative scores of work performance, quantitative scores of student satisfaction and quantitative scores of training participation of higher vocational counselors as the results of professional ability evaluation.

[0070] Step S103: construct a knowledge graph of recommended solutions for optimizing professional competence based on the professional competence assessment results and the results of the selection of improvement directions after self-assessment by the subject to be assessed.

[0071] Specifically, the Knowledge Graph is a modern theoretical tool that describes concepts, entities and their relationships in a structured semantic network.

[0072] Taking higher vocational counselors as an example, the results of the selection of directions for promotion after self-evaluation by higher vocational counselors are the results of the selection of directions for promotion after active self-evaluation by higher vocational counselors, so as to obtain the direction of professional ability optimized by self-selection of higher vocational counselors, and provide a basis for higher vocational counselors to actively optimize their own professional abilities based on subjective initiative.

[0073] Combining the professional ability assessment results and the results of the selection of the direction to be improved, a complete professional ability assessment result is obtained. Based on the complete professional ability assessment result, the entities and relationships are obtained, and each entity and relationship is converted into entity triples. Based on the entity triples, a knowledge graph of professional ability optimization recommendation solutions is constructed.

[0074] Step S104: obtaining a recommended professional capability optimization solution based on the professional capability optimization recommendation solution knowledge graph.

[0075] Specifically, the complete professional ability assessment results are matched with the knowledge graph of the recommended professional ability optimization plan for higher vocational counselors to obtain the recommended professional ability optimization plan for higher vocational counselors and the corresponding recommended professional ability optimization plan for higher vocational counselors.

[0076] The method for recommending a career optimization plan provided by this embodiment obtains the career ability data to be evaluated by the higher vocational counselors, and provides a basis for accurately learning the relationship between the career ability performance characteristics of the higher vocational counselors and the career ability evaluation results based on the comprehensive career ability performance data of the higher vocational counselors, so as to accurately predict the career ability level of the higher vocational counselors; based on the career ability to be evaluated data and the preset career ability evaluation model, the career ability of the subject to be evaluated is evaluated, and the career ability evaluation results are obtained; based on the career ability evaluation results and the results of the direction to be promoted after the self-evaluation of the subject to be evaluated, a knowledge graph of the career optimization recommendation plan is constructed; the results of the direction to be promoted obtained after the self-evaluation of the subject to be evaluated are obtained, so as to obtain the career to be self-selected and optimized by the higher vocational counselors. Ability direction provides a foundation for higher vocational counselors to actively optimize their own professional abilities based on subjective initiative. The knowledge graph of professional ability optimization recommendation plans, comprehensive professional ability assessment results and the results of selection of directions to be improved, are matched using the knowledge graph of professional ability optimization recommendation plans for higher vocational counselors to obtain the corresponding recommended professional ability optimization plans for higher vocational counselors; the recommended professional ability optimization plans based on the knowledge graph of professional ability optimization recommendation plans can accurately assess the professional ability level of higher vocational counselors and recommend their professional ability optimization plans in a targeted manner, so as to effectively improve the work ability and level of counselors, provide better services and support for students' growth and development, and solve the problem of lack of targeted recommendations for professional ability optimization of higher vocational counselors in existing technologies.

[0077] In this embodiment, a method for recommending a career capability optimization plan is provided, which can be used in the above-mentioned computer device. Figure 2 is a flow chart of a method for recommending a career capability optimization solution according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0078] Step S201: Obtain the professional competence data of the subject to be evaluated. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0079] In step S202, the data for professional competence to be evaluated include quantitative data of work performance, quantitative data of student satisfaction, and quantitative data of training participation. The quantitative data of work performance, quantitative data of student satisfaction, and quantitative data of training participation are subjected to z-score standardization respectively, and the quantitative data of work performance, quantitative data of student satisfaction, and quantitative data of training participation after z-score standardization are used as the data for professional competence to be evaluated.

[0080] Specifically, z-score normalization, also known as standard deviation normalization, is based on the mean and standard deviation of the original data, converting the data into a standard normal distribution with a mean of 0 and a standard deviation of 1. The calculation formula is:

[0081]

[0082] Where x is an observation in the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and z is the standardized data. This formula eliminates the effects of differences in dimensions and numerical ranges between different data, making different types of data comparable.

[0083] For example, we can collect quantitative data on the performance of vocational college counselors during their working period, such as the number of activities organized, the number of student awards, and the completion rate of work tasks. Let the set of these data be X = (x1, x2, ..., x n ), through the formula Calculate the mean of the original data by the formula Calculate the standard deviation of the original data, x i Represents the original data, and then performs data standardization according to formula (1), converting each original work performance data into the corresponding standardized data z i , and obtain the standardized work performance quantitative data set Z=(z1,z2,...,z n For example, if the original data of a counselor's task completion rate is 85%, the mean of the group's task completion rate is 75%, and the standard deviation is 5%, then the standardized value is (85-75) / 5=2.

[0084] Step S203 , performing a professional ability assessment on the subject to be assessed based on the professional ability assessment data and a preset professional ability assessment model to obtain a professional ability assessment result.

[0085] Specifically, the above step S203 includes:

[0086] In step S2031, the standardized work performance quantitative data, student satisfaction quantitative data and training participation quantitative data are respectively input into the preset professional ability assessment model to obtain the work performance quantitative score, student satisfaction quantitative score and training participation quantitative score.

[0087] In step S2032, the three-dimensional scores of the work performance quantitative score, the student satisfaction quantitative score and the training participation quantitative score are used as the professional ability assessment results.

[0088] Specifically, in this embodiment, taking the subject to be evaluated as a higher vocational counselor as an example, the construction and training method of the professional ability evaluation model includes the following steps:

[0089] Obtain the professional competence evaluation data of several higher vocational counselors during their working period, wherein the professional competence evaluation data of higher vocational counselors during their working period are the quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation after z-score standardization;

[0090] Three-dimensional professional ability scoring labels are set for each professional ability to be evaluated data, that is, the real category of professional ability of higher vocational counselors. The three-dimensional professional ability scoring labels refer to the quantitative scores of work performance, student satisfaction and training participation of higher vocational counselors. A professional ability evaluation dataset is constructed based on the professional ability to be evaluated data after setting the three-dimensional professional ability scoring labels.

[0091] Select the kernel function of the SVM model and set the penalty parameters and kernel function parameters.

[0092] The SVM model is trained according to the occupational ability assessment data set, and the optimal parameter combination is found through cross-validation, multiple separation surfaces are found, and a trained occupational ability assessment model is obtained; the trained occupational ability assessment model can accurately identify the three-dimensional occupational ability score corresponding to the input occupational ability data to be assessed, and use the corresponding three-dimensional occupational ability score as the occupational ability assessment result.

[0093] The result of the professional competence assessment is a three-dimensional score of professional competence consisting of the quantitative scores of work performance, student satisfaction and training participation of higher vocational counselors.

[0094] The quantitative scores for work performance include 9-10 points for excellent performance, 7-8 points for good performance, 5-6 points for average performance, and 1-4 points for unsatisfactory performance. Excellent performance corresponds to a college counselor who can efficiently complete various work tasks, has orderly student management, a positive class atmosphere, and significant student growth and progress. Good performance corresponds to a college counselor who can complete basic work tasks, but has deficiencies in student management and class atmosphere creation, and student growth and progress is not obvious. Average performance corresponds to a counselor who can complete basic work tasks, but has deficiencies in student management and class atmosphere creation, and student growth and progress is not obvious. Unsatisfactory performance corresponds to a counselor who often fails to complete tasks on time, has chaotic student management, a negative class atmosphere, and negatively affects student growth.

[0095] The quantitative scores for student satisfaction include very satisfied (9-10 points), satisfied (7-8 points), average (5-6 points), and dissatisfied (1-4 points). Very satisfied students believe that the counselors care about students, are responsible, and can provide help and support. Satisfied students believe that the counselors can basically meet students' needs and provide necessary help and support. Average students believe that the counselors have some shortcomings in some aspects, but are generally acceptable. Dissatisfied students believe that the counselors have obvious shortcomings in many aspects and cannot provide effective help and support.

[0096] The quantitative scores for training participation include 9-10 points for active participation, 7-8 points for good participation, 5-6 points for average participation, and 1-4 points for less participation; students who actively participate in the management of the corresponding counselors actively participate in various training activities, have a serious attitude towards learning, and take the initiative to ask questions and share experiences; students who participate well in the management of the corresponding counselors can participate in most training activities, have a relatively serious attitude towards learning, and can abide by training discipline; students who generally participate in the management of the corresponding counselors have average enthusiasm for participating in training activities, sometimes need to be urged to participate, and their learning attitude is not serious enough; students who rarely participate in the management of the corresponding counselors seldom participate in training activities, are often absent or late and leave early, and have an improper learning attitude.

[0097] Step S204: construct a knowledge graph of recommended solutions for optimizing professional competence based on the professional competence assessment results and the selection results of the direction for improvement after self-assessment by the subject to be assessed.

[0098] Specifically, taking the subject to be evaluated as a higher vocational counselor as an example, the above step S204 includes:

[0099] Step S2041, based on the professional ability assessment results, the results of the selection of the direction to be improved after the self-assessment of the assessed subject, and the preset personalized professional ability optimization plan, obtain the structured professional ability optimization plan data source and the unstructured professional ability optimization plan data source.

[0100] In step S2041, the vocational college counselor first obtains the results of the direction of improvement selected after self-assessment through the professional competence self-assessment multiple-choice questions. The professional competence self-assessment questions are questions for self-assessment of the direction of improvement of professional competence. The professional competence self-assessment multiple-choice questions include questions on professional background, interests and specialties, work experience, and students' actual needs. Questions on professional background include mental health education and career planning guidance; questions on interests and specialties include planning and implementing class team-building activities; questions on work experience include handling student attendance management, handling student violations and disciplinary violations, and guiding students to participate in scientific research projects; questions on students' actual needs include conducting psychological counseling and mental health lectures.

[0101] After summarizing the professional ability assessment results and the results of the selection of the direction to be improved, a complete professional ability assessment result is obtained.

[0102] In this embodiment, based on the management files of higher vocational counselors, the professional ability assessment results, the selection results of the direction to be promoted and the personalized professional ability optimization plan of each higher vocational counselor are obtained one by one, organized into structured data, and stored in a table of a relational database to obtain a structured higher vocational counselor professional ability optimization plan data source; a higher vocational counselor professional skills optimization questionnaire survey is conducted on the administrative management personnel of higher vocational counselors to obtain the professional ability assessment characteristics and professional ability optimization plans of different higher vocational counselors, and the professional ability assessment characteristics and professional ability optimization plans that are logically consistent are used as the higher vocational counselor professional ability optimization questionnaire survey results to form an unstructured higher vocational counselor professional ability optimization plan data source.

[0103] Step S2042: perform data cleaning on the structured professional capability optimization solution data source and the unstructured professional capability optimization solution data source respectively to remove duplicate, erroneous and inconsistent data, and obtain the stand-by structured professional capability optimization solution data source and the stand-by unstructured professional capability optimization solution data source respectively.

[0104] Specifically, data cleaning is performed on the structured and unstructured data sources for vocational counselors' professional competence optimization solutions, removing duplicate, erroneous, and inconsistent data. This results in a ready-to-use structured and unstructured data source for vocational counselors' professional competence optimization solutions. In step S2043, entity alignment and attribute filling are performed on the ready-to-use structured data source based on the preset management archive database and the preset professional competence optimization solution database to obtain knowledge fusion data.

[0105] Specifically, based on the higher vocational counselor management archive database and the preset professional ability optimization program database, the structured higher vocational counselor professional ability optimization program data source to be used is entity aligned and attribute filled to obtain knowledge fusion data.

[0106] Furthermore, entity alignment can be achieved by:

[0107] Rule-based alignment: Define entity alignment rules, such as matching based on unique identifiers such as solution names and numbers. Use Python's pandas library for data matching and the merge() function to associate data based on specified columns.

[0108] Machine learning-based alignment: Utilize entity linking algorithms, such as similarity-based methods (cosine similarity, edit distance, etc.) or deep learning models (such as BERT's semantic matching model), to automatically identify records representing the same entity in different data sources.

[0109] Attribute filling: Supplementary information is obtained from the preset management archive database and the professional competency optimization plan database to fill in missing attributes in the structured data. For example, if an optimization plan lacks applicable population information, the corresponding information can be obtained from the management archive database by linking the plan number and filling it in using an UPDATE statement.

[0110] Step S2044: perform entity extraction, relationship extraction, and attribute extraction on the unstructured professional capability optimization solution data source to obtain knowledge extraction data.

[0111] In an optional implementation, the above step S2044 includes:

[0112] Step a1, identify and locate entities in the data source of the unstructured professional ability optimization plan to be used, and classify the entities into predefined entity categories as entity extraction data, where the predefined entity categories include personalized professional ability optimization plans, professional ability assessment results, and results of selection of directions to be improved.

[0113] Specifically, the unstructured data source for the professional competency optimization solution is cleaned twice, using regular expressions to remove noise data such as special symbols and HTML tags. Long texts are segmented into short sentences to facilitate model processing.

[0114] Using annotation tools like Prodigy or brat, organize annotation personnel to annotate text according to predefined entity categories (personalized professional competency optimization plan, professional competency assessment results, and results of potential improvement areas). Clarify annotation standards during annotation. For example, "personalized professional competency optimization plan" requires the full plan name and key description to ensure consistency. Data augmentation is performed on the annotated data, expanding the dataset through methods such as synonym replacement and sentence order adjustment.

[0115] Use BiLSTM-CRF or Transformer-based models (such as BERT and RoBERTa) for entity extraction. Convert the labeled data into a format acceptable to the model and set training parameters (such as learning rate, batch size, and number of training rounds). Use PyTorch or TensorFlow frameworks for training, recording the loss function value and accuracy during training, and adjusting hyperparameters using the validation set to prevent overfitting. Input the preprocessed unstructured data into the trained model for prediction, outputting the entity label for each token. Post-process the prediction results, such as removing overlapping annotations and merging adjacent entities of the same category, to obtain the entity extraction data.

[0116] Step a2, based on statistical learning technology, uses deep learning algorithms to learn the characteristics of the relationship between entities, and uses the learned characteristics to extract the relationship between entities from the unstructured professional ability optimization solution data source to be used as relationship extraction data.

[0117] Specifically, we focus on the entity and extract several words before and after it as contextual features. In Python, we can obtain context through string slicing. We use NLTK (Natural Language Toolkit) or spaCy (a natural language processing toolkit) to perform part-of-speech tagging on the text and extract part-of-speech information of the entity and its context as features.

[0118] Dependency parsing is performed using tools such as Stanford CoreNLP (Natural Language Processing Toolkit) or AllenNLP. Syntactic relationships between entities (such as subject-verb and verb-object) are extracted as features to construct a structured relational representation. The extracted features are combined with labeled entity relationship data to form training samples. A relation extraction model based on CNN (Convolutional Neural Network) or GNN (Graph Neural Network) is used. Using GNN as an example, graph-structured data is constructed, with entities as nodes and connections between entities as edges. Relational features between nodes are learned through graph convolution operations. The cross-entropy loss function is used as the optimization objective, and optimizers such as Adam and Adagrad are selected for model training. Early stopping is used during training to prevent overfitting, and the optimal number of training rounds is determined based on performance indicators on the validation set.

[0119] The processed unstructured data is fed into the trained model to predict relationships between entities. The model outputs a probability distribution for the relationships between each entity pair and selects the relationship with the highest probability as the prediction. A confidence threshold is set to filter out relationship predictions with probabilities below the threshold, retaining the extracted data with high confidence.

[0120] In step a3, the attributes of each entity are extracted from the unstructured professional capability optimization solution data source using the attribute extraction deep learning network as attribute extraction data.

[0121] Specifically, in this embodiment, the deep learning network is trained and predicted based on the professional ability optimization plan for higher vocational counselors. The trained deep learning network is used as an attribute extraction deep learning network to achieve the attribute extraction task. The deep learning network adopts a convolutional neural network.

[0122] The text containing the entity is used as the input of the convolutional neural network model, and the output is the attribute information of the entity. Design the attribute template and clarify the output format, such as the key-value pair format of "attribute name: attribute value". When labeling the entity, label the relevant attribute information of the entity. For example, for the "Personalized Professional Competence Optimization Plan" entity, label its "applicable population", "implementation cycle", "expected effect" and other attributes. Convert the labeled attribute data into an input and output format acceptable to the model, such as encoding the text and attribute information into a token sequence, and adding special tags to distinguish different attributes. Set the training parameters and use the AdamW optimizer and learning rate scheduler (such as cosine annealing learning rate) for training. During the training process, evaluate the model performance through the validation set and adjust the hyperparameters to optimize the model effect. Input the preprocessed unstructured data into the trained model for inference, and the model outputs the attribute information of the entity. Parse and format the output results to obtain the final attribute extraction data.

[0123] Step a4: Use the entity extraction data, relationship extraction data and attribute extraction data as knowledge extraction data.

[0124] Specifically, the entity extraction data, relationship extraction data, and attribute extraction data are integrated to form a unified knowledge extraction data structure. For example, it is stored in JSON format, organizing each entity and its corresponding relationship and attribute information into an object.

[0125] The method for recommending a professional capability optimization solution provided in this embodiment can quickly sort out key information in the data by identifying and locating entities in unstructured data sources and dividing them into predefined entity categories, such as personalized professional capability optimization solutions, professional capability assessment results, and results of selecting directions for improvement. This targeted entity classification method lays a clear framework foundation for the subsequent construction of a knowledge graph. By comprehensively applying statistical learning techniques and deep learning algorithms, the relationships between entities are extracted from unstructured data, so that the knowledge graph can more comprehensively and accurately reflect the logical connections between the various elements in the process of professional capability optimization, providing strong support for accurately recommending optimization solutions. Entity attributes are extracted using an attribute extraction deep learning network. The network has powerful feature learning capabilities and can accurately extract detailed attribute information of each entity from complex unstructured text. Integrating entity extraction, relationship extraction, and attribute extraction into a complete data processing process enables efficient conversion from unstructured data sources to knowledge extraction data, which can greatly enhance the practicality of the knowledge graph for recommending professional capability optimization solutions.

[0126] Step S2045 , performing coreference resolution and entity disambiguation on the knowledge fusion data and the knowledge extraction data to obtain the relationships between entities, and constructing an initial database based on the relationships between entities.

[0127] Specifically, coreference resolution refers to the use of clustering algorithms (such as hierarchical clustering, DBSCAN) or coreference resolution models based on deep learning (such as the End-to-End model) to identify different expressions in the text that refer to the same entity and merge them into the same entity.

[0128] Entity disambiguation involves mapping ambiguous entities to precise entities within the knowledge base through entity linking. For example, for the entity "communication skills," context can be used to determine whether it specifically refers to "counselor-student communication skills" or "teamwork communication skills." The processed entities and their relationships are stored in an initial database, either using a graph database (such as Neo4j) or a relational database (using foreign key associations to simulate a graph structure).

[0129] Step S2046: Convert the entities and relationships of the initial database into entity triples in the database, and construct a knowledge graph of professional capability optimization recommendation solutions based on the entity triples.

[0130] Specifically, the entities and relationships in the initial database are converted into entity triplets (head entity, relationship, tail entity). Knowledge graph construction tools (such as Neo4j's Cypher language and Dgraph's GraphQL interface) are used to import these triples and construct a knowledge graph for recommending professional competency optimization solutions. For graph visualization, tools such as Gephi and Cytoscape can be used to visualize the graph, allowing users to intuitively view entity relationships and knowledge structures.

[0131] Step S205: Get the recommended professional ability optimization solution based on the professional ability optimization recommendation solution knowledge graph. Figure 1 Step S104 of the illustrated embodiment will not be described in detail here.

[0132] The method for recommending a career capability optimization plan provided in this embodiment integrates the career capability assessment results, the selection results of the direction to be promoted after the self-assessment of the subject to be assessed, and the structured and unstructured data sources of the preset personalized career capability optimization plan, breaking down data barriers and integrating multi-channel information. Data cleaning is performed on the structured and unstructured data sources respectively to remove duplicate, erroneous and inconsistent data. This effectively avoids errors or inaccuracies in the knowledge graph caused by data quality problems, ensures that the knowledge graph constructed subsequently is based on high-quality data, performs entity alignment and attribute filling on the structured data source to be used, and extracts entities, relationships and attributes on the unstructured data source to be used, converting scattered data into structured knowledge. Through knowledge fusion and extraction, the potential correlation between the data is excavated, so that the knowledge graph can clearly present the logical connection between the career capability assessment results, the direction to be promoted and the optimization plan, and performs coreference resolution and entity disambiguation on the knowledge fusion data and the knowledge extraction data to solve the problem of multiple expressions or ambiguity of the same entity, ensuring the uniqueness and accuracy of the entities and relationships in the knowledge graph. The initial database's entities and relationships are converted into triples of each entity within the database to construct a knowledge graph. This structured storage facilitates knowledge querying, updating, and maintenance. Furthermore, the triple-based knowledge graph structure is easily integrated with other systems, enabling rapid and efficient recommendation of appropriate optimization solutions based on professional competency assessment results and areas for improvement.

[0133] In this embodiment, a method for recommending a career capability optimization plan is provided, which can be used in the above-mentioned computer device. Figure 3 is a flow chart of a method for recommending a career capability optimization solution according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0134] Step S301: Obtain the professional competence data to be assessed. Figure 2 Step S201 of the illustrated embodiment will not be described in detail here.

[0135] Step S302: Based on the professional ability assessment data and the preset professional ability assessment model, the professional ability assessment of the subject to be assessed is performed to obtain the professional ability assessment result. Figure 2 Step S203 of the illustrated embodiment will not be described in detail here.

[0136] Step S303: Construct a knowledge graph of recommended solutions for optimizing professional competence based on the professional competence assessment results and the self-assessment results of the candidates for improvement. Figure 2 Step S204 of the illustrated embodiment will not be described in detail here.

[0137] Step S304: obtaining a recommended professional capability optimization solution based on the professional capability optimization recommendation solution knowledge graph.

[0138] According to the complete professional ability assessment results, based on the knowledge graph of recommended solutions for optimizing the professional ability of higher vocational counselors, a recommended solution for optimizing the professional ability of higher vocational counselors is obtained.

[0139] Specifically, taking the subject to be evaluated as a higher vocational counselor as an example, the above step S304 includes:

[0140] Step S3041: obtain a set of objects to be evaluated, set a bipartite graph between each object to be evaluated in the set and the professional capability optimization solution, and obtain the interaction relationship between the object to be evaluated and the professional capability optimization solution based on the bipartite graph.

[0141] Specifically, define the bipartite graph of vocational counselors and professional ability optimization plan:

[0142] G={(p,y pi ,i)|p∈P,i∈I} (2);

[0143] Among them, G represents the bipartite graph of vocational counselors and professional ability optimization schemes, p represents vocational counselors, y pi represents the interactive relationship between higher vocational counselors and the professional ability optimization plan, i represents the professional ability optimization plan, ∈ represents belonging, P represents the set of higher vocational counselors, I represents the set of professional ability optimization plans, |· represents constraints, wherein the interactive relationship is to input the complete professional ability assessment results; in this embodiment, by inputting the complete professional ability assessment results, it is possible to accurately recommend the professional ability optimization plan for the higher vocational counselor based on the professional ability assessment results of the higher vocational counselor and the selection results of the direction to be promoted.

[0144] Step S3042: Add the set of objects to be evaluated and the interaction relationships to the knowledge graph of the professional capability optimization recommendation plan to form an attention network of the professional capability optimization knowledge graph; the attention network of the professional capability optimization knowledge graph includes nodes and edges.

[0145] Specifically, the set of objects to be evaluated and the interactive relationship data are integrated with the constructed knowledge graph of the professional ability optimization recommendation plan. If a graph database (such as Neo4j) is used to store the knowledge graph, new nodes and edges can be added to the graph through Cypher statements. The Graph Attention Network (GAT) structure is implemented using a deep learning framework (such as PyTorch-Geometric). Node features and edge features are defined. Node features can include the professional ability evaluation results and basic information of the object to be evaluated, as well as attribute information of the professional ability optimization plan. Through the multi-head attention mechanism, the model pays attention to different aspects of node and edge information, calculates the attention weights between nodes, and achieves focus on key information.

[0146] Step S3043: Use the preset network knowledge representation model to learn the vector representation of nodes and edges in the attention network of the professional ability optimization knowledge graph to obtain the first-order embedded vector representation of the object to be evaluated and the professional ability optimization plan.

[0147] Specifically, the calculation expression of the preset network knowledge representation model is as follows:

[0148]

[0149] Among them, f r (h, t) represents the projection score of the network knowledge representation model, h represents the head entity, t represents the tail entity, represents the mapped head entity vector, r represents the relationship between entities, represents the tail entity vector after mapping, Represents the square operation of the two norm, W r Represents the projection matrix from entity space to relational space. The network knowledge representation model provided in this embodiment can effectively handle many-to-many relationships by mapping entities into different relational spaces, while also being highly flexible and easy to train. The rationality of a triple can be measured using the projection score of the knowledge representation model: the rationality of a triple is evaluated by calculating the distance between the sum of the head entity vector and the relation vector projected into the relational space and the tail entity vector. A lower score indicates a more reasonable triple, while a higher score indicates a less reasonable triple.

[0150] Step S3044, based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan, the preset attention embedding propagation module is used to update the node representation in the attention network of the professional ability optimization knowledge graph, and generate an aggregated attention vector representation of multiple objects to be evaluated and professional ability optimization plans. The preset attention embedding propagation module includes several propagation layers, and each propagation layer updates the node representation by recursively propagating the embedding vector of the neighboring node.

[0151] Specifically, in this embodiment, the attention embedding propagation module determines the degree of influence of each node on its neighboring nodes through the similarity between adjacent nodes, and each node updates its embedding vector representation according to the embedding vector of its neighboring node and the corresponding attention score weight, thereby establishing an association relationship between nodes; the representations output by different propagation layers emphasize the connectivity information of different orders, which can effectively enrich the embedding vector representation of higher vocational counselors and professional capacity optimization plans.

[0152] In some optional implementations, step S3044 includes:

[0153] Step b1: construct a set of head entity triples.

[0154] The formula for the entity triple set is as follows:

[0155] N h ={(h,r,t)|(h,r,t)∈G} (5);

[0156] Among them, N h Represents a set of head entity triples.

[0157] Step b2: obtaining the information content of the tail entity vector of the head entity triplet based on the head entity triplet set.

[0158] Specifically, the formula for obtaining the information content of the tail entity vector of the head entity triplet based on the head entity triplet set is as follows:

[0159]

[0160] Among them, e Nh represents the amount of information in the tail entity vector of the head entity triple, π(h, r, t) represents the attention weight, and e t Represents the tail entity vector, e h Represents the head entity vector, e r Represents the inter-entity relationship vector; in this embodiment, the attention degree weight can represent the amount of information transferred from the tail entity vector to the head entity vector based on the inter-entity relationship vector.

[0161] In step b3, based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan and the information content of the tail entity vector of the head entity triplet, the representations of the object to be evaluated and the professional ability optimization plan output by all propagation layers are aggregated to obtain the aggregated attention vector representation of the object to be evaluated and the professional ability optimization plan.

[0162] Specifically, the calculation expression of the dual interaction aggregator is as follows:

[0163] f BiIn =LeakyRelu(W1(e h +e Nh ))+LeakyRelu(W2(e h ⊙e Nh ))(8);

[0164] Among them, f BiInDenotes a dual interaction aggregator, LeakyRelu(·) denotes a LeakyRelu activation function, W1 denotes a first trainable weight matrix, W2 denotes a second trainable weight matrix, and ⊙ denotes a vector element-wise product operation. This embodiment aggregates all propagation layers of vocational counselors and professional competence optimization solutions, allowing the propagated information to focus more on the correlation between the information content of the head entity vector and the tail entity vector of the head entity triple, thereby transferring more information between entities.

[0165] Step S3045: Concatenate the first-order embedding vector representation of the object to be evaluated and the professional capability optimization plan, the node representation of each propagation layer, and the aggregated attention vector representation to obtain a matching score for each object to be evaluated and the professional capability optimization plan.

[0166] Specifically, the calculation expression for the matching score between higher vocational counselors and the professional ability optimization plan is as follows:

[0167]

[0168] in, represents the matching score between higher vocational counselors and the career capability optimization plan, represents the target after splicing, T represents transposition, The target occupational ability optimization plan after splicing is expressed as follows: Represents the first-order embedding vector representation of vocational counselors, represents the representation of the vocational counselor output by the first propagation layer, || represents splicing, represents the aggregated attention vector representation of the vocational counselor, The first-order embedding vector representation of the professional ability optimization plan, Representation of the professional ability optimization solution output by the first propagation layer, The aggregated attention vector representation of the professional capability optimization scheme is represented, and L represents the total number of propagation layers.

[0169] Step S3046: The occupational capability optimization plan corresponding to the object to be evaluated when the matching score is the highest is used as the recommended occupational capability optimization plan.

[0170] Specifically, all matching scores of each object to be evaluated are sorted, and the occupational capability optimization plan with the highest score is selected as the recommendation result, and fed back to the object to be evaluated to complete the recommendation process of the occupational capability optimization plan.

[0171] The method for recommending a professional capability optimization solution provided in this embodiment presents complex correspondences in an intuitive and clear graphical structure by constructing a bipartite graph between the object to be evaluated and the professional capability optimization solution. This method can quickly locate the association between the object to be evaluated and the optimization solution, forming an attention network of the professional capability optimization knowledge graph. By utilizing the characteristics of the attention mechanism, it can automatically focus on important nodes and edges in the knowledge graph that are related to the recommendation of professional capability optimization solutions. By using a preset network knowledge representation model to learn the vector representations of nodes and edges, and converting the object to be evaluated and the professional capability optimization solution into a low-dimensional dense vector form, it is possible to effectively mine the potential features and semantic information behind the data. These vector representations not only contain the explicit attributes of the entity, but also capture the complex implicit relationships between entities. The preset attention embedding propagation module updates the node representation by recursively propagating the embedding vectors of neighboring nodes through multiple layers. The propagation of each layer is a further fusion and refinement of knowledge. This layered propagation mechanism aggregates information at varying depths and breadths, allowing node representations to continuously absorb knowledge from surrounding related nodes, thereby more comprehensively reflecting the combined characteristics of the evaluated object and the optimization solution. It comprehensively considers the first-order embedding vector representation, the node representations of each propagation layer, and the aggregated attention vector representation to calculate the matching score from multiple dimensions. This multi-dimensional evaluation method comprehensively measures the fit between the evaluated object and the professional competency optimization solution, avoiding the one-sidedness of single-dimensional evaluation. Ultimately, the solution with the highest matching score is recommended, significantly improving the adaptability of the recommended solution to the actual needs of the evaluated object and achieving precise recommendations.

[0172] As one or more specific application embodiments of the present invention, combined with Figure 4 The method for recommending a professional capability optimization solution provided by the present invention is further described in detail. Figure 4 As shown, the specific process includes:

[0173] Step S1, obtaining quantitative data on work performance, student satisfaction and training participation of vocational college counselors during their work; wherein, the quantitative data on work performance include the effectiveness of student academic guidance, the effectiveness of academic style construction and the effectiveness of student affairs management; the quantitative data on student satisfaction include the student satisfaction assessment score and the student evaluation ratio; the quantitative data on training participation include the number of training participants, the training participation rate, the number and quality of training courses.

[0174] Step S2: Perform z-score standardization on the quantitative data of work performance, quantitative data of student satisfaction, and quantitative data of training participation of higher vocational counselors during their work, and use the standardized data as the data to be evaluated for professional ability.

[0175] Step S3: Based on the occupational ability to be assessed data, the trained occupational ability assessment model is used to perform occupational ability assessment to obtain the occupational ability assessment result.

[0176] In this embodiment, the construction and training method of the professional ability assessment model includes the following steps:

[0177] The professional ability evaluation data of several higher vocational counselors during their work period were obtained, among which the professional ability evaluation data of higher vocational counselors during their work period were the quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation after z-score standardization.

[0178] Three-dimensional professional ability scoring labels are set for each professional ability to be evaluated data, that is, the real category of professional ability of higher vocational counselors. The three-dimensional professional ability scoring labels refer to the quantitative scores of work performance, student satisfaction and training participation of higher vocational counselors. A professional ability evaluation dataset is constructed based on the professional ability to be evaluated data after setting the three-dimensional professional ability scoring labels.

[0179] Select the kernel function of the SVM model and set the penalty parameters and kernel function parameters.

[0180] The SVM model is trained according to the occupational ability assessment data set, and the optimal parameter combination is found through cross-validation, multiple separation surfaces are found, and a trained occupational ability assessment model is obtained; the trained occupational ability assessment model can accurately identify the three-dimensional occupational ability score corresponding to the input occupational ability data to be assessed, and use the corresponding three-dimensional occupational ability score as the occupational ability assessment result.

[0181] The professional competence assessment result in step S3 is a three-dimensional professional competence score consisting of the quantitative scores of the work performance, student satisfaction and training participation of the higher vocational counselors.

[0182] The quantitative scores for work performance include 9-10 points for excellent performance, 7-8 points for good performance, 5-6 points for average performance, and 1-4 points for unsatisfactory performance; excellent performance corresponds to higher vocational counselors who can efficiently complete various work tasks, have orderly student management, a positive class atmosphere, and significant student growth and progress; good performance corresponds to higher vocational counselors who can complete basic work tasks, but have deficiencies in student management and class atmosphere creation, and student growth and progress are not obvious; average performance corresponds to counselors who can complete basic work tasks, but have deficiencies in student management and class atmosphere creation, and student growth and progress are not obvious; unsatisfactory performance corresponds to counselors who are often unable to complete tasks on time, have chaotic student management, a negative class atmosphere, and negatively affected student growth.

[0183] The quantitative scores of student satisfaction include very satisfied (9-10 points), satisfied (7-8 points), average (5-6 points), and dissatisfied (1-4 points); very satisfied students believe that the counselors care about students, are responsible, and can provide help and support; satisfied students believe that the counselors can basically meet students' needs and provide necessary help and support; average students believe that the counselors have shortcomings in some aspects but are generally acceptable; dissatisfied students believe that the counselors have obvious shortcomings in many aspects and cannot provide effective help and support.

[0184] The quantitative scores for training participation include 9-10 points for active participation, 7-8 points for good participation, 5-6 points for average participation, and 1-4 points for less participation; students who actively participate in the management of the corresponding counselors actively participate in various training activities, have a serious attitude towards learning, and take the initiative to ask questions and share experiences; students who participate well in the management of the corresponding counselors can participate in most training activities, have a relatively serious attitude towards learning, and can abide by training discipline; students who generally participate in the management of the corresponding counselors have average enthusiasm for participating in training activities, sometimes need to be urged to participate, and their learning attitude is not serious enough; students who rarely participate in the management of the corresponding counselors seldom participate in training activities, are often absent or late and leave early, and have an improper learning attitude.

[0185] Step S4: obtaining the results of the vocational college counselors' selection of the direction to be improved after self-assessment through the vocational ability self-assessment multiple-choice questions; the vocational ability self-assessment questions are questions for selecting the direction to be improved of the self-assessment vocational ability;

[0186] The multiple-choice questions on the self-assessment of professional ability include questions on professional background, interests and specialties, work experience and students' actual needs; questions on professional background include mental health education, career planning guidance, etc.; questions on interests and specialties include the planning and implementation of class team-building activities, etc.; questions on work experience include handling student attendance management, handling student violations of rules and regulations, guiding students to participate in scientific research projects, etc.; questions on students' actual needs include conducting psychological counseling and mental health lectures, etc.

[0187] Step S5: Combine the professional ability assessment results and the results of the selection of the direction to be promoted to obtain a complete professional ability assessment result.

[0188] Step S6: Construct a knowledge graph of recommended solutions for optimizing the professional capabilities of higher vocational counselors.

[0189] Step S6 includes the following steps:

[0190] Step S61, obtain the structured data source of the vocational college counselor's professional ability optimization plan and the unstructured data source of the vocational college counselor's professional ability optimization plan; in this embodiment, according to the vocational college counselor management file, obtain the vocational ability assessment results, the selection results of the direction to be promoted and the personalized vocational ability optimization plan of each vocational college counselor one by one, organize them into structured data, and store them in a table in a relational database to obtain the structured data source of the vocational college counselor's professional ability optimization plan; conduct a vocational college counselor professional skills optimization questionnaire survey on the administrative management personnel of the vocational college counselors to obtain the vocational ability assessment characteristics and vocational ability optimization plans of different vocational college counselors, and use the vocational ability assessment characteristics and vocational ability optimization plans that are in line with logical consistency as the vocational college counselor professional ability optimization questionnaire survey results to form an unstructured data source of the vocational college counselor's professional ability optimization plan.

[0191] Step S62: perform data cleaning on the structured higher vocational counselor professional ability optimization program data source and the unstructured higher vocational counselor professional ability optimization program data source respectively to remove duplicate, erroneous and inconsistent data, and obtain the stand-by structured higher vocational counselor professional ability optimization program data source and the stand-by unstructured higher vocational counselor professional ability optimization program data source.

[0192] Step S63: Based on the higher vocational counselor management file database and the professional ability optimization program database, entity alignment and attribute filling are performed on the structured higher vocational counselor professional ability optimization program data source to be used to obtain knowledge fusion data.

[0193] Step S64: perform entity extraction, relationship extraction, and attribute extraction on the unstructured data source of the vocational college counselor professional ability optimization solution to obtain knowledge extraction data.

[0194] Step S64 includes the following steps:

[0195] Step S641: Identify and locate entities in the unstructured data source of the vocational counselor professional competence optimization plan to be used, and classify the entities into predefined entity categories as entity extraction data, wherein the predefined entity categories include vocational competence optimization plans, vocational competence assessment results, and results of selection of directions to be promoted;

[0196] Step S642: Using a deep learning algorithm to learn features of relationships between entities based on a statistical learning method, and using the learned features to extract relationships between entities from a data source of a stand-by unstructured vocational counselor professional competence optimization solution as relationship extraction data;

[0197] Step S643: Utilizing the attribute extraction deep learning network, extracting the attributes of each entity from the unstructured data source of the vocational counselor professional competence optimization plan to be used as attribute extraction data;

[0198] In this embodiment, a deep learning network is trained and predicted based on the vocational ability optimization plan for higher vocational counselors. The trained deep learning network is used as an attribute extraction deep learning network to achieve the attribute extraction task. The deep learning network adopts a convolutional neural network.

[0199] Step S644: Use the entity extraction data, relationship extraction data and attribute extraction data as knowledge extraction data.

[0200] Step S65: perform coreference resolution and entity disambiguation on the knowledge fusion data and the knowledge extraction data, and construct an initial database.

[0201] Step S66: Convert the entities and relationships in the initial database into entity triples in the Neo4j database to complete the construction of the knowledge graph of the recommended solution for optimizing the professional competence of higher vocational counselors.

[0202] Step S7: According to the complete professional ability assessment results, based on the knowledge graph of recommended professional ability optimization plans for higher vocational counselors, a recommended professional ability optimization plan for higher vocational counselors is obtained.

[0203] Step S7 includes the following steps:

[0204] Step S71: define a bipartite graph of vocational counselors and the vocational ability optimization plan:

[0205] G={(p,y pi ,i)|p∈P,i∈I} (2).

[0206] Step S72: Add the set of higher vocational counselors and their interaction relationships to the knowledge graph of the recommended plan for optimizing the professional ability of higher vocational counselors to form an attention network of the knowledge graph for optimizing the professional ability.

[0207] Step S73: Use the network knowledge representation model to learn the vector representations of nodes and edges in the attention network of the vocational ability optimization knowledge graph, and obtain the first-order embedded vector representation of the higher vocational counselors and the vocational ability optimization plan.

[0208] The calculation expression of the network knowledge representation model in step S73 is as follows:

[0209]

[0210] Step S74: Based on the first-order embedding vector representation of higher vocational counselors and professional ability optimization solutions, the attention embedding propagation module is used to update the node representation in the attention network of the professional ability optimization knowledge graph to obtain the aggregated attention vector representation of several higher vocational counselors and professional ability optimization solutions, wherein the attention embedding propagation module includes several propagation layers, and each propagation layer updates the node representation by recursively propagating the embedding vector of the neighboring node; in this embodiment, the attention embedding propagation module determines the degree of influence of each node on its neighboring node through the similarity between adjacent nodes, and each node updates its embedding vector representation according to the embedding vector of its neighboring node and the corresponding attention score weight, thereby establishing an association relationship between nodes; the representations output by different propagation layers emphasize the connectivity information of different orders, which can effectively enrich the embedding vector representation of higher vocational counselors and professional ability optimization solutions.

[0211] Step S74 includes the following steps:

[0212] Step S741: Construct a header entity triple set. The formula of the entity triple set is as follows:

[0213] N h ={(h,r,t)|(h,r,t)∈G} (5).

[0214] Step S742: According to the head entity triple set, the information content of the tail entity vector of the head entity triple is obtained:

[0215]

[0216] Step S743: Based on the first-order embedding vector representations of the vocational counselors and the professional competence optimization solutions, and based on the information content of the tail entity vectors of the head entity triples, the representations of the vocational counselors and the professional competence optimization solutions output by all propagation layers are aggregated to obtain the aggregated attention vector representations of the vocational counselors and the professional competence optimization solutions. In step S743, a dual-interaction aggregator is used to aggregate the representations of the vocational counselors and the professional competence optimization solutions output by each propagation layer;

[0217] The calculation expression of the dual interaction aggregator is as follows:

[0218] f BiIn =LeakyRelu(W1(e h +e Nh ))+LeakyRelu(W2(e h ⊙e Nh ))(8).

[0219] Step S75: Concatenate the first-order embedding vector representations of each vocational counselor and the vocational ability optimization plan, the representations output by each propagation layer, and the aggregated attention vector representation to obtain the matching score between the vocational counselor and the vocational ability optimization plan. The calculation expression for the matching score between the vocational counselor and the vocational ability optimization plan in step S75 is as follows:

[0220]

[0221] Step S76: The vocational ability optimization plan corresponding to the higher vocational counselor with the highest matching score is used as the recommended vocational ability optimization plan.

[0222] The method for recommending a career optimization plan provided by this embodiment obtains quantitative data on work performance, quantitative data on student satisfaction and quantitative data on training participation of higher vocational counselors during their work, extracts corresponding quantitative data on career capabilities from three aspects: work assessment of higher vocational counselors, student satisfaction and student management training, as data for career capability to be evaluated, thereby providing a basis for accurately learning the relationship between the career capability performance characteristics of higher vocational counselors and the results of career capability evaluation based on the comprehensive career capability performance data of higher vocational counselors, so as to accurately predict the corresponding career capability level of higher vocational counselors; the present invention realizes the evaluation of comprehensive career capability evaluation results based on the quantitative data of the three dimensions of work performance, student satisfaction and training participation of higher vocational counselors through a trained career capability evaluation model; the present invention obtains the data for evaluation by obtaining the results for evaluation by higher vocational counselors after self-evaluation Improve the direction selection results to obtain the self-selected and optimized professional ability direction of the higher vocational counselors, and provide a basis for the higher vocational counselors to actively optimize their own professional abilities based on their subjective initiative; the present invention obtains a complete professional ability evaluation result by combining the professional ability evaluation results and the selection results of the direction to be improved, and constructs a knowledge map of the recommended program for optimizing the professional ability of higher vocational counselors, and realizes matching based on the professional ability evaluation results and the selection results of the direction to be improved in the complete professional ability evaluation results, using the knowledge map of the recommended program for optimizing the professional ability of higher vocational counselors to obtain the corresponding recommended professional ability optimization program for higher vocational counselors; the present invention can accurately evaluate the professional ability level of higher vocational counselors and recommend their professional ability optimization programs in a targeted manner, so as to effectively improve the work ability and level of counselors, and provide better services and support for the growth and development of students.

[0223] In this embodiment, a device for recommending a career optimization plan is also provided. The device is used to implement the above-mentioned embodiments and preferred embodiments, and the details that have been described will not be repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0224] This embodiment provides a device for recommending a career capability optimization plan. Figure 5 As shown, including:

[0225] The module 501 for obtaining data to be evaluated is used to obtain the data to be evaluated on the professional ability of the object to be evaluated.

[0226] The professional ability assessment module 502 is used to perform professional ability assessment on the subject to be assessed based on the professional ability assessment data and a preset professional ability assessment model to obtain a professional ability assessment result.

[0227] The knowledge graph construction module 503 is used to construct a knowledge graph of the career capability optimization recommendation plan based on the career capability assessment results and the results of the improvement direction selection after the self-assessment of the assessed subject.

[0228] The professional capability optimization plan recommendation module 504 is used to obtain recommended professional capability optimization plans based on the professional capability optimization recommendation plan knowledge graph.

[0229] In some optional implementations, the professional competence assessment data includes quantitative work performance data, quantitative student satisfaction data, and quantitative training participation data, and the professional competence optimization solution recommendation device further includes:

[0230] The z-score standardization processing module is used to perform z-score standardization on the work performance quantitative data, student satisfaction quantitative data and training participation quantitative data respectively, and use the z-score standardized work performance quantitative data, student satisfaction quantitative data and training participation quantitative data as the data to be evaluated for professional ability.

[0231] In some optional implementations, the professional competency assessment module 502 includes:

[0232] The professional competency assessment unit is used to input the standardized work performance quantitative data, student satisfaction quantitative data and training participation quantitative data into the preset professional competency assessment model to obtain the work performance quantitative score, student satisfaction quantitative score and training participation quantitative score.

[0233] The professional competency assessment result determination unit is used to use the three-dimensional scores of work performance quantitative scores, student satisfaction quantitative scores and training participation quantitative scores as the professional competency assessment results.

[0234] In some optional implementations, the knowledge graph construction module 503 includes:

[0235] The data source acquisition unit is used to obtain the structured professional ability optimization plan data source and the unstructured professional ability optimization plan data source based on the professional ability assessment results, the selection results of the direction to be improved after self-assessment of the assessed subject, and the preset personalized professional ability optimization plan.

[0236] The data source cleaning unit is used to clean the structured professional capability optimization plan data source and the unstructured professional capability optimization plan data source respectively, remove duplicate, erroneous and inconsistent data, and obtain the stand-by structured professional capability optimization plan data source and the stand-by unstructured professional capability optimization plan data source respectively.

[0237] The knowledge fusion data determination unit is used to perform entity alignment and attribute filling on the structured professional capability optimization program data source to be used based on the preset management file database and the preset professional capability optimization program database to obtain knowledge fusion data.

[0238] The knowledge extraction unit is used to extract entities, relationships and attributes from the unstructured professional capability optimization solution data source to obtain knowledge extraction data.

[0239] The entity relationship determination unit is used to perform coreference resolution and entity disambiguation on the knowledge fusion data and the knowledge extraction data to obtain the relationship between entities, and to build an initial database based on the relationship between entities.

[0240] The knowledge graph construction unit is used to convert the entities and relationships of the initial database into entity triples in the database, and to construct a knowledge graph of professional ability optimization recommendation solutions based on the entity triples.

[0241] In some optional implementations, the knowledge extraction unit includes:

[0242] The entity extraction sub-unit is used to identify and locate entities in the data source of the unstructured professional ability optimization plan to be used, and classify the entities into predefined entity categories as entity extraction data, where the predefined entity categories include personalized professional ability optimization plans, professional ability assessment results and results of selection of directions to be improved.

[0243] The relationship extraction subunit is used to learn the characteristics of the relationship between entities using a deep learning algorithm based on statistical learning technology, and use the learned characteristics to extract the relationship between entities from the unstructured professional ability optimization solution data source as relationship extraction data.

[0244] The attribute extraction subunit is used to extract the attributes of each entity from the unstructured professional ability optimization solution data source using the attribute extraction deep learning network as attribute extraction data.

[0245] The knowledge extraction data integration unit is used to use entity extraction data, relationship extraction data and attribute extraction data as knowledge extraction data.

[0246] In some optional implementations, the professional capability optimization solution recommendation module 504 includes:

[0247] The interactive relationship determination unit is used to obtain a set of objects to be evaluated, set a bipartite graph between each object to be evaluated in the set of objects to be evaluated and the professional ability optimization plan, and obtain the interactive relationship between the object to be evaluated and the professional ability optimization plan based on the bipartite graph.

[0248] The attention network generation unit is used to add the set of objects to be evaluated and the interaction relationship to the knowledge graph of the professional ability optimization recommendation plan to form an attention network of the professional ability optimization knowledge graph; the attention network of the professional ability optimization knowledge graph includes nodes and edges.

[0249] The first-order embedding vector representation generation unit is used to use the preset network knowledge representation model to learn the vector representation of nodes and edges in the attention network of the professional ability optimization knowledge graph, and obtain the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan.

[0250] The aggregated attention vector representation generation unit is used to update the node representation in the attention network of the professional ability optimization knowledge graph based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan using a preset attention embedding propagation module, and generate the aggregated attention vector representation of multiple objects to be evaluated and professional ability optimization plans. The preset attention embedding propagation module includes several propagation layers, and each propagation layer updates the node representation by recursively propagating the embedding vector of the neighboring node.

[0251] The matching unit is used to splice the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan, the node representation of each propagation layer, and the aggregated attention vector representation to obtain the matching score of each object to be evaluated and the professional ability optimization plan.

[0252] The professional ability optimization plan recommendation unit is used to recommend the professional ability optimization plan corresponding to the object to be evaluated when the matching score is the highest as the recommended professional ability optimization plan.

[0253] In some optional embodiments, the aggregate attention vector representation generation unit includes:

[0254] The head entity triple set construction subunit is used to construct the head entity triple set.

[0255] An information quantity generating subunit, configured to obtain the information quantity of the tail entity vector of the head entity triple based on the head entity triple set;

[0256] The aggregated attention vector representation generation subunit is used to aggregate the representations of the object to be evaluated and the professional ability optimization plan output by all propagation layers based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan and the information content of the tail entity vector of the head entity triplet, and obtain the aggregated attention vector representation of the object to be evaluated and the professional ability optimization plan.

[0257] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0258] The career capability optimization program recommendation device in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0259] The embodiment of the present invention also provides a computer device having the above Figure 5 The recommended device for the occupational capability optimization program shown.

[0260] See also Figure 6 , Figure 6 is a structural diagram of a computer device provided by an optional embodiment of the present invention, such as Figure 6 As shown, the computer device includes: one or more processors 10, memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components utilize different buses to communicate with each other and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in the memory or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Equally, multiple computer devices can be connected, and each device provides part of the necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 6 A processor 10 is taken as an example.

[0261] The processor 10 may be a central processing unit, a network processor, or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic, or any combination thereof.

[0262] The memory 20 stores instructions that can be executed by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiment.

[0263] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and application programs required for at least one function; the data storage area may store data created based on the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely located relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0264] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0265] The computer device further includes an input device 30 and an output device 40. The processor 10, the memory 20, the input device 30 and the output device 40 may be connected via a bus or other means. Figure 6 The bus connection is taken as an example.

[0266] The input device 30 can receive input digital or character information and generate key signal input related to user settings and function control of the computer device, such as a touch screen, a keypad, a mouse, a trackpad, a touch pad, an indicator stick, one or more mouse buttons, a trackball, a joystick, etc. The output device 40 can include a display device, an auxiliary lighting device (e.g., an LED), and a tactile feedback device (e.g., a vibration motor). The above-mentioned display device includes but is not limited to a liquid crystal display, a light emitting diode, a display, and a plasma display. In some optional embodiments, the display device can be a touch screen.

[0267] The embodiment of the present invention also provides a computer-readable storage medium. The above-mentioned method according to the embodiment of the present invention can be implemented in hardware, firmware, or implemented as a computer code that can be recorded in a storage medium, or implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memory. It can be understood that a computer, a processor, a microprocessor controller or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0268] A portion of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the form in which the computer program instruction exists in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc. Accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium that can be accessed by the computer.

[0269] Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention. Such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for recommending a professional capability optimization plan, characterized in that: The method comprises: Obtain the professional competence assessment data of the subject to be assessed; Performing a professional competency assessment on the subject to be assessed based on the professional competency assessment data and a preset professional competency assessment model to obtain a professional competency assessment result; Constructing a knowledge graph of recommended solutions for optimizing professional competence based on the professional competence assessment results and the selection results of the direction to be improved after the self-assessment of the assessed subjects; A recommended career capability optimization plan is obtained based on the career capability optimization recommendation plan knowledge graph.

2. The method according to claim 1, characterized in that The professional competence data to be assessed include quantitative work performance data, quantitative student satisfaction data, and quantitative training participation data. Before performing the professional competence assessment on the subject to be assessed based on the professional competence data to be assessed and a preset professional competence assessment model, the method further includes: The quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation were subjected to z-score standardization respectively, and the quantitative data of work performance, quantitative data of student satisfaction and quantitative data of training participation after z-score standardization were used as the data to be evaluated for professional ability.

3. The method according to claim 2, characterized in that The professional ability assessment of the subject to be assessed is performed based on the professional ability assessment data and the preset professional ability assessment model to obtain a professional ability assessment result, including: Inputting the standardized work performance quantitative data, student satisfaction quantitative data, and training participation quantitative data into a preset professional competency assessment model to obtain a work performance quantitative score, a student satisfaction quantitative score, and a training participation quantitative score; The three-dimensional scores of work performance quantitative scores, student satisfaction quantitative scores and training participation quantitative scores are used as the results of professional ability assessment.

4. The method according to claim 1, wherein The construction of a knowledge graph of a professional competency optimization recommendation solution based on the professional competency assessment results and the selection results of the direction to be improved after the self-assessment of the assessed subject includes: Obtaining a structured career capability optimization plan data source and an unstructured career capability optimization plan data source based on the career capability assessment results, the results of the promotion direction selection after the self-assessment of the subject to be assessed, and the preset personalized career capability optimization plan; Performing data cleaning on the structured professional capability optimization solution data source and the unstructured professional capability optimization solution data source respectively to remove duplicate, erroneous and inconsistent data, thereby obtaining a stand-by structured professional capability optimization solution data source and a stand-by unstructured professional capability optimization solution data source respectively; Based on a preset management archive database and a preset professional capability optimization solution database, entity alignment and attribute filling are performed on the data source of the structured professional capability optimization solution to be used to obtain knowledge fusion data; Performing entity extraction, relationship extraction, and attribute extraction on the unstructured professional capability optimization solution data source to obtain knowledge extraction data; Performing coreference resolution and entity disambiguation on the knowledge fusion data and the knowledge extraction data to obtain relationships between entities, and constructing an initial database based on the relationships between the entities; The entities and relationships of the initial database are converted into entity triples in the database, and a knowledge graph of professional ability optimization recommendation solutions is constructed based on the entity triples.

5. The method according to claim 4, characterized in that Entity extraction, relationship extraction, and attribute extraction are performed on the unstructured professional capability optimization solution data source to obtain knowledge extraction data, including: Identify and locate entities in the data source of the unstructured professional competence optimization plan to be used, and classify the entities into predefined entity categories as entity extraction data. The predefined entity categories include personalized professional competence optimization plans, professional competence assessment results, and results of selection of areas for improvement. Based on statistical learning technology, deep learning algorithms are used to learn the characteristics of the relationship between entities, and the learned characteristics are used to extract the relationship between entities from the unstructured professional ability optimization solution data source to be used as relationship extraction data; Utilizing the attribute extraction deep learning network, the attributes of each entity are extracted from the unstructured occupational capability optimization solution data source as attribute extraction data; The entity extraction data, the relationship extraction data and the attribute extraction data are used as knowledge extraction data.

6. The method according to claim 1, characterized in that The recommended professional capability optimization solution based on the professional capability optimization recommendation solution knowledge graph includes: Obtaining a set of objects to be evaluated, setting a bipartite graph between each object to be evaluated in the set of objects to be evaluated and the professional capability optimization solution, and obtaining an interactive relationship between the object to be evaluated and the professional capability optimization solution based on the bipartite graph; Adding the set of objects to be evaluated and their interaction relationships to the knowledge graph of the professional capability optimization recommendation solution to form an attention network of the professional capability optimization knowledge graph; the attention network of the professional capability optimization knowledge graph includes nodes and edges; The preset network knowledge representation model is used to learn the vector representations of nodes and edges in the attention network of the professional ability optimization knowledge graph, and the first-order embedding vector representations of the object to be evaluated and the professional ability optimization solution are obtained; Based on the first-order embedding vector representations of the objects to be evaluated and the professional ability optimization solutions, a preset attention embedding propagation module is used to update the node representations in the attention network of the professional ability optimization knowledge graph, and generate aggregated attention vector representations of multiple objects to be evaluated and professional ability optimization solutions. The preset attention embedding propagation module includes several propagation layers, each of which recursively propagates the embedding vectors of neighboring nodes to update the node representation; Concatenate the first-order embedding vector representation of the object to be evaluated and the professional ability optimization solution, the node representation of each propagation layer, and the aggregated attention vector representation to obtain a matching score for each object to be evaluated and the professional ability optimization solution; The occupational capability optimization plan corresponding to the object to be evaluated when the matching score is the highest will be used as the recommended occupational capability optimization plan.

7. The method according to claim 6, characterized in that Based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan, the preset attention embedding propagation module is used to update the node representation in the attention network of the professional ability optimization knowledge graph, and generate the aggregated attention vector representation of multiple objects to be evaluated and professional ability optimization plans, including: Construct a set of head entity triples; Based on the head entity triple set, the information content of the tail entity vector of the head entity triple is obtained; Based on the first-order embedding vector representation of the object to be evaluated and the professional ability optimization plan and the information content of the tail entity vector of the head entity triplet, the representations of the object to be evaluated and the professional ability optimization plan output by all propagation layers are aggregated respectively to obtain the aggregated attention vector representation of the object to be evaluated and the professional ability optimization plan.

8. A device for recommending a professional ability optimization plan, characterized in that: The device comprises: The module for obtaining data to be evaluated is used to obtain the data on the professional ability to be evaluated of the object to be evaluated; A professional competency assessment module is used to perform a professional competency assessment on the subject to be assessed based on the professional competency assessment data and a preset professional competency assessment model to obtain a professional competency assessment result; A knowledge graph construction module is used to construct a knowledge graph of a professional ability optimization recommendation plan based on the professional ability assessment results and the results of the improvement direction selected by the subject to be assessed after self-assessment; The professional ability optimization plan recommendation module is used to obtain recommended professional ability optimization plans based on the professional ability optimization recommendation plan knowledge graph.

9. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, computer instructions are stored in the memory, and the processor executes the method for recommending a professional capability optimization plan according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method for recommending a professional capability optimization plan according to any one of claims 1 to 7.

Images