Artificial intelligence field talent training and evaluation system

Abstract

The invention relates to the technical field of training, in particular to a talent training and evaluation system in the field of artificial intelligence, and the system comprises a model construction module which is used for constructing a target object training recommendation model and an evaluation model; the calculation module is used for calculating an optimal training strategy suitable for the target object; the test module is used for testing the target object by using the evaluation model; and the determination module is used for determining talent information based on the test result of the evaluation model on the target object. According to the method, the personalized optimal training strategy is generated by constructing the target object training recommendation model, so that unnecessary resource waste is reduced; the evaluation model can objectively and accurately evaluate the learning achievement and ability level of the target object, and the accuracy and fairness of evaluation are improved; the talent information base converts the evaluation result into detailed talent information, so that organizations or enterprises can know the structure and distribution of internal talents more clearly, and powerful support is provided for talent selection, cultivation and incentive.

Description

Technical Field

[0001] The present invention relates to the field of training technology, and in particular to a talent training and assessment system in the field of artificial intelligence. Background Art

[0002] With the rapid development of science and technology, artificial intelligence (AI) technology has gradually penetrated into various fields of social economy and has become an important force in promoting industrial upgrading and economic development. However, the rapid development of artificial intelligence technology has also brought about a sharp increase in the demand for professional talents. At present, there is a shortage of talents in the field of artificial intelligence worldwide, especially the shortage of high-end talents is becoming increasingly prominent.

[0003] The traditional talent training and assessment system is unable to cope with the needs of the field of artificial intelligence. On the one hand, the traditional education system is relatively single in talent training in the field of artificial intelligence and cannot provide training based on the actual situation of the target objects. On the other hand, traditional talent assessment methods often rely on manual operation and subjective judgment, and have problems such as low efficiency and lack of accuracy, which cannot meet the needs of the field of artificial intelligence for comprehensive and accurate talent assessment. Summary of the invention

[0004] 1. Purpose of the invention The purpose of the present invention is to provide a talent training and assessment system in the field of artificial intelligence that can improve the accuracy of target object training and meet the needs of comprehensive and accurate assessment of talents in the field of artificial intelligence.

[0005] (II) Technical solution To solve the above problems, the present invention provides a talent training and assessment system in the field of artificial intelligence, comprising: Model building module, which is used to build target object training recommendation model and evaluation model; An information collection module, which is used to collect information about the target object and input it into the training recommendation model; A calculation module, which is used to calculate the optimal training strategy suitable for the target object according to the training recommendation model; A testing module, which is used to test the target object using the evaluation model; A determination module is used to determine talent information based on the test results of the target object by the evaluation model and store it in a talent information database, wherein the talent information at least includes talent gradient information.

[0006] In another aspect of the present invention, preferably, the target object's information includes professional information and course information; The training recommendation model is: ; in, is the information of the input target object; TF (P, C) represents the talent feature extraction function; P represents the professional information; C represents the course information; RC represents the training content recommendation function; KG represents the knowledge graph in the field of artificial intelligence; Calculating the optimal training strategy suitable for the target object according to the training recommendation model includes: Calculate the talent feature extraction function based on the professional information feature extraction function and the course information feature extraction function The value of Extract function based on calculated talent characteristics The value and knowledge graph A recommended list of training content is determined as the optimal training strategy.

[0007] In another aspect of the present invention, preferably, the talent feature extraction function for: ; in, Represented as professional information feature extraction function, Represented as course information feature extraction function.

[0008] In another aspect of the present invention, preferably, the professional information feature extraction function for: ; Among them, m represents the professional characteristic dimension; p j represents the professional characteristics of talents in the jth dimension; u j represents the mean professional characteristics of all talents in the jth dimension; represents the variance of professional characteristics of all talents in the jth dimension; The course information feature extraction function for: ; Where l represents the number of courses; c k represents the mastery level of the talent in the kth course; w k represents the weight of the kth course.

[0009] In another aspect of the present invention, preferably, the function of the optimal training strategy is: ; Among them, S represents the set of training content in the knowledge graph, r s It represents the relevance weight of the training content s; TF represents the talent feature extraction function; KG represents the knowledge graph in the field of artificial intelligence.

[0010] In another aspect of the present invention, preferably, testing the target object using the evaluation model includes testing using the following formula: ; Among them, D represents the test results of the evaluation model, γ1 represents the weight coefficient of the performance evaluation sub-model, D1 represents the test results of the performance evaluation sub-model, γ2 represents the weight coefficient of the learning quality evaluation sub-model, D2 represents the test results of the learning quality evaluation sub-model, γ3 represents the weight coefficient of the personality evaluation sub-model, and D3 represents the test results of the personality evaluation sub-model.

[0011] In another aspect of the present invention, preferably, the performance evaluation sub-model is calculated using the following formula: ; Among them, D1 represents the test result of the performance evaluation sub-model, y represents the yth test, and Y represents the total number of tests. represents the yth test score of the i-th trainee in the same period, N represents the total number of trainees in the same period, It represents the y-1th test score of the i-th trainee in the same period.

[0012] In another aspect of the present invention, preferably, the learning quality assessment sub-model is calculated using the following formula: ; Among them, D2 represents the test result of the learning quality evaluation sub-model. Indicates the maximum timeliness of job completion. Indicates the minimum timeliness of job completion. represents the mean value of timeliness of job completion, represents the mean time of interaction with the teacher, and a represents the number of interactions with the teacher.

[0013] In another aspect of the present invention, preferably, the personality assessment sub-model includes personality assessment based on dominance, influence, stability and compliance.

[0014] In another aspect of the present invention, preferably, determining talent information based on the result of testing the target object with the evaluation model includes: If the test results of the performance evaluation sub-model, the test results of the learning quality evaluation sub-model, and the test results of the personality evaluation sub-model are all greater than or equal to the first score threshold, the target object is determined as the first talent echelon; If a portion of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is greater than or equal to the first score threshold but another portion is greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the second talent echelon; If the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model are all greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the third talent echelon; If at least one of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is less than the second score threshold, the target object is determined as the fourth talent echelon; The first score threshold is greater than the second score threshold.

[0015] (III) Beneficial effects The above technical solution of the present invention has the following beneficial technical effects: The present invention can generate personalized optimal training strategies for each target object's unique background and ability level by constructing a target object training recommendation model. This personalized training program can significantly improve the training effect, help trainees improve their skill level faster, and reduce unnecessary waste of resources; the evaluation model can objectively and accurately evaluate the learning outcomes and ability level of the target object, and provide a scientific basis for subsequent training and career development planning. This efficient evaluation mechanism not only saves manpower and material resources, but also improves the accuracy and fairness of the evaluation. The talent information database of the present invention converts the evaluation results into detailed talent information, which helps organizations or enterprises to more clearly understand the structure and distribution of internal talents, and provides strong support for talent selection, training and motivation. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic diagram of the overall structure of an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0018] Obviously, the described embodiments are only some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In the description of the present invention, it should be noted that the terms “first”, “second” and “third” are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0020] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0021] Embodiment 1 A talent training and assessment system in the field of artificial intelligence. Figure 1 FIG. 1 shows a schematic diagram of the overall structure of an embodiment of the present invention. Figure 1 As shown, including: The model building module is used to build a target object training recommendation model and an evaluation model; the target object training recommendation model in this embodiment is:

[0022] in, is the input information of the target object; TF (P, C) represents the talent feature extraction function; P is the professional information; C is the course information, TF (P, C) represents the talent feature extraction function, which is used to extract key features from the professional information (P) and course information (C) of the target object to construct the feature vector of the target object; RC represents the training content recommendation function, which recommends the most suitable training content for the target object based on the feature vector extracted by TF function and the knowledge graph (KG) in the field of artificial intelligence; KG represents the knowledge graph in the field of artificial intelligence, which contains various knowledge, skills and their relationships in this field, and is an important reference for the recommendation system.

[0023] The information collection module is used to collect the information of the target object and input it into the training recommendation model; in this embodiment, the information of the target object includes professional information and course information. The professional information may involve educational background, professional direction, work experience, etc.; the course information records the courses that the target object has learned or is learning and the degree of mastery. For example, Zhang San is a graduate of computer science and hopes to improve his professional skills in the field of artificial intelligence. The information of the target object Zhang San includes professional information P: professional skills and ability scores: programming (Python, Java), data structure, algorithm, database, etc., programming ability (90%), mathematical foundation (80%), machine learning knowledge (70%), research experience (60%); course information C: courses learned and mastery: basic machine learning (80%), introduction to deep learning (70%), natural language processing (60%).

[0024] A calculation module, which is used to calculate the optimal training strategy suitable for the target object according to the training recommendation model, including: Calculate the talent feature extraction function based on the professional information feature extraction function and the course information feature extraction function The value of Extract function based on calculated talent characteristics The value and knowledge graph A recommended list of training content is determined as the optimal training strategy.

[0025] In this embodiment, the talent feature extraction function for: ; in, Represented as professional information feature extraction function, Represented as course information feature extraction function.

[0026] Professional information feature extraction function for: ; Among them, m represents the professional characteristic dimension; p j represents the professional characteristics of talents in the jth dimension; u j represents the mean professional characteristics of all talents in the jth dimension; Represents the variance of the professional characteristics of all talents in the jth dimension; through standardization, professional characteristics of different dimensions and distributions are converted into comparable values ​​to facilitate subsequent analysis.

[0027] Assume that the knowledge graph has the following professional characteristic dimensions: programming ability, mathematical foundation, machine learning knowledge, and research experience.

[0028] According to Zhang San's professional information, it is assumed that the following eigenvector p can be obtained: ; Calculate the mean u and variance , obtained through statistics of a large amount of talent data; ; Apply the formula to calculate : ; Course information feature extraction function for: ; Where l represents the number of courses; c k represents the mastery level of the talent in the kth course; w k represents the weight of the kth course, reflecting the importance of the course in the field of artificial intelligence. According to Zhang San’s course mastery, the following eigenvector c can be obtained: ; Assume that the weight w of each course is as follows: ; Apply the formula to calculate : ; Calculate the talent feature TF, and Combined to calculate talent characteristics TF:

[0029] The function of the optimal training strategy is: ; Among them, S represents the set of training content in the knowledge graph, r s represents the relevance weight of training content s; TF represents the talent feature extraction function; KG represents the knowledge graph in the field of artificial intelligence. Assume that the knowledge graph KG contains the following training content set S: Advanced Machine Learning, Deep Learning Applications, Advanced Natural Language Processing, and Basic Computer Vision; each training content s has a relevance weight r s , here we assume that r s It is calculated based on the correlation between the content in the knowledge graph and the talent characteristics; all permutations and combinations of advanced machine learning, deep learning applications, advanced natural language processing, and basic computer vision, such as two-by-two combinations, three-way combinations, etc., are calculated to obtain the maximum value of the combination, and the combination corresponding to the maximum value is the optimal training strategy.

[0030] Assume that the relevance weight r of machine learning is advanced s =0.9, relevance weight r of deep learning applications s =0.8, natural language processing advanced relevance weight r s =0.7, computer vision basis relevance weight r s =0.6; When there are four combinations, the calculated value is: , the calculated values ​​for the three combinations are as follows: Etc., the calculated values ​​are not listed one by one here, and the calculated values ​​of all combinations are compared. The combination corresponding to the maximum value is the optimal training strategy.

[0031] A testing module is used to test the target object using the evaluation model. In this embodiment, testing the target object using the evaluation model includes testing using the following formula: ; Where D represents the test result of the evaluation model, γ1 represents the weight coefficient of the performance evaluation sub-model, D1 represents the test result of the performance evaluation sub-model, γ2 represents the weight coefficient of the learning quality evaluation sub-model, D2 represents the test result of the learning quality evaluation sub-model, γ3 represents the weight coefficient of the personality evaluation sub-model, and D3 represents the test result of the personality evaluation sub-model; The performance evaluation sub-model is calculated using the following formula: ; Among them, D1 represents the test result of the performance evaluation sub-model, y represents the yth test, and Y represents the total number of tests. represents the yth test score of the i-th trainee in the same period, N represents the total number of trainees in the same period, It represents the y-1th test score of the i-th trainee in the same period.

[0032] The learning quality evaluation sub-model is calculated using the following formula: ; Among them, D2 represents the test result of the learning quality evaluation sub-model. Indicates the maximum timeliness of job completion. Indicates the minimum timeliness of job completion. represents the mean value of timeliness of job completion, represents the mean time of interaction with the teacher, and a represents the number of interactions with the teacher.

[0033] It should be noted that the above-mentioned "teacher" refers to an entity that imparts knowledge, skills or values, which includes both human individuals in the biological sense and systematic teaching assistance platforms built based on artificial intelligence, computer technology, etc. The "teacher" has the following characteristics: first, it has teaching ability and can provide corresponding teaching content and methods according to the needs of learners; second, it can interact to guide, coach and evaluate learners; third, it can adapt to different learning scenarios, including but not limited to face-to-face teaching, distance education, and self-help learning environments.

[0034] The personality assessment sub-model includes personality assessment based on dominance, influence, stability and compliance, calculated using the following formula: ; Among them: D3 represents the test result of the personality assessment sub-model, z1 represents dominance, ranging from [0, 100], indicating the dominance of the target object; z2 represents influence, ranging from [0, 100], indicating the target object's ability to influence others; z3 represents stability, ranging from [0, 100], indicating the target object's tendency to work stably in the long term; z4 represents obedience, ranging from [0, 100], indicating the target object's tendency to obey arrangements.

[0035] A determination module is used to determine talent information based on the test results of the target object by the evaluation model and store it in a talent information database, wherein the talent information at least includes talent gradient information, including: If the test results of the performance evaluation sub-model, the test results of the learning quality evaluation sub-model, and the test results of the personality evaluation sub-model are all greater than or equal to the first score threshold, the target object is determined as the first talent echelon; If a portion of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is greater than or equal to the first score threshold but another portion is greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the second talent echelon; If the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model are all greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the third talent echelon; If at least one of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is less than the second score threshold, the target object is determined as the fourth talent echelon; The first score threshold is greater than the second score threshold.

[0036] The present invention can generate personalized optimal training strategies for each target object's unique background and ability level by constructing a target object training recommendation model. This personalized training program can significantly improve the training effect, help trainees improve their skill level faster, and reduce unnecessary waste of resources; the evaluation model can objectively and accurately evaluate the learning outcomes and ability level of the target object, and provide a scientific basis for subsequent training and career development planning. This efficient evaluation mechanism not only saves manpower and material resources, but also improves the accuracy and fairness of the evaluation. The talent information database of the present invention converts the evaluation results into detailed talent information, which helps organizations or enterprises to more clearly understand the structure and distribution of internal talents, and provides strong support for talent selection, training and motivation.

[0037] It should be understood that the above specific embodiments of the present invention are only used to illustrate or explain the principles of the present invention, and do not constitute a limitation of the present invention. Therefore, any modifications, equivalent substitutions, improvements, etc. made without departing from the spirit and scope of the present invention should be included in the protection scope of the present invention. In addition, the appended claims of the present invention are intended to cover all changes and modifications that fall within the scope and boundaries of the appended claims, or the equivalent forms of such scope and boundaries.

[0038] The present invention has been described above with reference to the embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. The scope of the present invention is defined by the appended claims and their equivalents. Without departing from the scope of the present invention, a person skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

[0039] Although the embodiments of the present invention have been described in detail, it should be understood that the various changes, substitutions, and alterations could be made hereto without departing from the spirit and scope of the invention.

[0040] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A talent training and assessment system in the field of artificial intelligence, characterized in that: include: Model building module, which is used to build target object training recommendation model and evaluation model; An information collection module, which is used to collect information about the target object and input it into the training recommendation model; A calculation module, which is used to calculate the optimal training strategy suitable for the target object according to the training recommendation model; A testing module, which is used to test the target object using the evaluation model; A determination module is used to determine talent information based on the test results of the target object by the evaluation model and store it in a talent information database, wherein the talent information at least includes talent gradient information.

2. The talent training and assessment system in the field of artificial intelligence according to claim 1 is characterized in that: The target object's information includes professional information and course information; The training recommendation model is: ； in, is the information of the input target object; TF (P, C) represents the talent feature extraction function; P represents the professional information; C represents the course information; RC represents the training content recommendation function; KG represents the knowledge graph in the field of artificial intelligence; Calculating the optimal training strategy suitable for the target object according to the training recommendation model includes: Calculate the talent feature extraction function based on the professional information feature extraction function and the course information feature extraction function The value of Extract function based on calculated talent characteristics The value and knowledge graph A recommended list of training content is determined as the optimal training strategy.

3. The talent training and assessment system in the field of artificial intelligence according to claim 2 is characterized in that: The talent feature extraction function for: ； in, Represented as professional information feature extraction function, Represented as course information feature extraction function.

4. The talent training and assessment system in the field of artificial intelligence according to claim 3 is characterized in that: The professional information feature extraction function for: ； Among them, m represents the professional characteristic dimension; p j represents the professional characteristics of talents in the jth dimension; u j represents the mean professional characteristics of all talents in the jth dimension; represents the variance of professional characteristics of all talents in the jth dimension; The course information feature extraction function for: ； Where l represents the number of courses; c k represents the mastery level of the talent in the kth course; w k represents the weight of the kth course.

5. The talent training and assessment system in the field of artificial intelligence according to claim 3 is characterized in that: The function of the optimal training strategy is: ； Among them, S represents the set of training content in the knowledge graph, r s It represents the relevance weight of the training content s; TF represents the talent feature extraction function; KG represents the knowledge graph in the field of artificial intelligence.

6. The talent training and assessment system in the field of artificial intelligence according to claim 1, characterized in that: Testing the target object using the evaluation model includes testing using the following formula: ； Among them, D represents the test results of the evaluation model, γ1 represents the weight coefficient of the performance evaluation sub-model, D1 represents the test results of the performance evaluation sub-model, γ2 represents the weight coefficient of the learning quality evaluation sub-model, D2 represents the test results of the learning quality evaluation sub-model, γ3 represents the weight coefficient of the personality evaluation sub-model, and D3 represents the test results of the personality evaluation sub-model.

7. The talent training and assessment system in the field of artificial intelligence according to claim 6, characterized in that: The performance evaluation sub-model is calculated using the following formula: ； Among them, D1 represents the test result of the performance evaluation sub-model, y represents the yth test, and Y represents the total number of tests. represents the yth test score of the i-th trainee in the same period, N represents the total number of trainees in the same period, It represents the y-1th test score of the i-th trainee in the same period.

8. The artificial intelligence talent training and assessment system according to claim 6, characterized in that: The learning quality evaluation sub-model is calculated using the following formula: ； Among them, D2 represents the test result of the learning quality evaluation sub-model. Indicates the maximum timeliness of job completion. Indicates the minimum timeliness of job completion. represents the mean value of timeliness of job completion, represents the mean time of interaction with the teacher, and a represents the number of interactions with the teacher.

9. The talent training and assessment system in the field of artificial intelligence according to claim 6, characterized in that: The personality assessment sub-model includes personality assessment based on dominance, influence, stability and compliance.

10. The artificial intelligence talent training and assessment system according to claim 6, characterized in that: Determining talent information based on the results of testing the target object with the assessment model includes: If the test results of the performance evaluation sub-model, the test results of the learning quality evaluation sub-model, and the test results of the personality evaluation sub-model are all greater than or equal to the first score threshold, the target object is determined as the first talent echelon; If a portion of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is greater than or equal to the first score threshold but another portion is greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the second talent echelon; If the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model are all greater than or equal to the second score threshold and less than the first score threshold, the target object is determined as the third talent echelon; If at least one of the test results of the performance assessment sub-model, the test results of the learning quality assessment sub-model, and the test results of the personality assessment sub-model is less than the second score threshold, the target object is determined as the fourth talent echelon; The first score threshold is greater than the second score threshold.

Images