Evaluation assisting method and system based on artificial intelligence

Abstract

The invention discloses an evaluation assisting method and system based on artificial intelligence, and relates to the field of education evaluation, and the method comprises an education evaluation assisting center which comprises an intelligent collection module, an intelligent analysis module, an intelligent storage module, a self-adaptive management module and an evaluation analysis module; the intelligent acquisition module acquires education knowledge information, education evaluation information and user behavior data; the intelligent analysis module constructs an evaluation knowledge point evaluation vector space according to the educational knowledge information; the intelligent storage module matches the education evaluation information with corresponding education knowledge information in the evaluation knowledge point evaluation vector space to generate an evaluation knowledge storage space; the self-adaptive management module performs personalized evaluation according to the user behavior data and generates personalized evaluation feedback information; the evaluation analysis module dynamically feeds the education evaluation information in the evaluation knowledge storage space back to the user evaluation process according to the personalized evaluation feedback information; according to the method, errors caused by personalized differences in the evaluation process are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of educational evaluation, and particularly to an evaluation assistance method and system based on artificial intelligence. Background Art

[0002] With the development of information technology, some evaluation assistance technologies have emerged. Online evaluation systems have improved the efficiency and convenience of evaluation to a certain extent, and can achieve automated question distribution and scoring. In recent years, artificial intelligence technologies have achieved rapid development, bringing new opportunities and challenges to the evaluation field. After retrieval, the invention patent with the Chinese patent number CN119228221A discloses an educational evaluation assistance method and system based on artificial intelligence. The method includes data collection, data optimization, learning behavior evaluation, educational feature evaluation, and educational evaluation assistance. The present invention relates to the technical field of educational management, specifically referring to an educational evaluation assistance method and system based on artificial intelligence. This solution adopts a two-way correlation analysis method combining learning behavior and educational features, conducts preliminary analysis through learning behavior evaluation, and refines educational features from multiple dimensions through educational feature multi-dimensional extraction, improving the analysis depth and specificity of educational evaluation. At the same time, it provides an effective way for intelligent methods to assist educational evaluation. The method of combining a graph neural long short-term memory integration model with causal heuristic enhancement is used for learning behavior evaluation. The method of combining a graph neural network of knowledge transfer and knowledge graph is used for educational feature evaluation.

[0003] Compared with the prior art, the invention patent with the Chinese patent number CN119228221A can refine educational features from multiple dimensions through a two-way correlation analysis method combining learning behavior and educational features. In addition, by constructing an educational feature knowledge graph to supplement educational feature information in other dimensions, it not only improves the analysis depth and specificity of educational evaluation, but also improves the operation efficiency and expandability of the evaluation.

[0004] However, in the actual use process of the above method, there are individual differences in the teaching of teachers and the learning of students, and it is impossible to accurately use the same educational feature information to assist in evaluating different users. Due to ignoring individual differences, the accuracy in the process of educational evaluation is reduced to a certain extent. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcoming of reduced accuracy caused by individual differences in the prior art, and to propose an evaluation assistance method and system based on artificial intelligence.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions: An evaluation assistance system based on artificial intelligence, including an education evaluation assistance center, which includes an intelligent acquisition module, an intelligent analysis module, an intelligent storage module, an adaptive management module, and an evaluation analysis module; The intelligent acquisition module is used to collect educational knowledge information, educational evaluation information, and user behavior data during the user evaluation process; The intelligent analysis module is used to systematically analyze the collected educational knowledge information, obtain corresponding evaluation knowledge point vectors, and construct an evaluation knowledge point evaluation vector space based on the evaluation knowledge point vectors; The intelligent storage module is used to perform matching analysis on educational evaluation information and corresponding educational knowledge information, and plan and store the corresponding educational evaluation information in the evaluation knowledge evaluation vector space according to the matching analysis result to generate an evaluation knowledge storage space; The adaptive management module is used to perform personalized evaluation on the evaluation results and evaluation process of the knowledge feature points corresponding to the user behavior data and educational evaluation information, and obtain personalized evaluation feedback information corresponding to the user; The evaluation analysis module is used to obtain the relevant educational evaluation information stored in the evaluation knowledge storage space according to the personalized evaluation feedback information, dynamically feedback the educational evaluation information to the user evaluation process, and obtain the educational evaluation results of the corresponding user.

[0007] The above technical solution further includes: The process of collecting educational knowledge information, educational evaluation information, and user behavior data during the user evaluation process includes: Set an evaluation management window, which is used for corresponding educational evaluation staff to enter basic educational evaluation information. The basic educational evaluation information includes corresponding educational evaluation scope information, educational evaluation target information, and educational evaluation informed information; Set multiple information retrieval sub-windows according to the educational evaluation scope information, and obtain educational knowledge information and educational evaluation information through the corresponding information retrieval sub-windows; Obtain the corresponding user basic information during the educational test evaluation process, monitor the test evaluation process of the corresponding user through intelligent acquisition devices, and obtain the corresponding user behavior data. The user behavior data includes evaluation answering process data and evaluation answering result data Furthermore, the process of obtaining evaluation knowledge point vectors includes: Obtain educational knowledge information, which includes educational planning information and educational knowledge point information. Extract knowledge features from the knowledge point definition information corresponding to the educational knowledge point information to obtain corresponding knowledge feature points; Based on natural language processing technology, obtain the association relationship between the corresponding knowledge feature points, and construct a knowledge point feature graph structure according to the corresponding knowledge feature points and association relationships; Analyze and train the corresponding knowledge point feature map structure based on the graph neural network algorithm, output the feature vectors corresponding to each knowledge feature point in the corresponding education knowledge point information, obtain the feature vector set corresponding to the corresponding education knowledge point information, and integrate and process each feature vector in the corresponding feature vector set to obtain the evaluation knowledge point vector corresponding to the corresponding education knowledge point information.

[0008] Furthermore, the process of constructing the evaluation knowledge point evaluation vector space includes: According to the education planning information, preliminarily arrange and distribute the evaluation knowledge point vectors corresponding to the corresponding education knowledge point information to construct an initial distribution vector space; Conduct a similarity correlation analysis on the evaluation knowledge point vectors corresponding to each education knowledge point information in the initial distribution vector space, sequentially set corresponding vector correlation control groups for each evaluation knowledge point vector, and split and process the evaluation knowledge point vectors in the vector correlation control groups according to the corresponding integration process to obtain the feature vectors corresponding to each knowledge feature point in the feature vector set; Obtain a feature vector correlation subset according to the corresponding feature vectors, conduct a correlation analysis on the feature vectors in the feature vector correlation subset, and determine whether there may be a correlation in the corresponding feature vector correlation subset; Analyze and process the feature vector correlation subsets that may be correlated in the vector correlation control group, and obtain the control similarity data CT of the corresponding vector correlation control group through the following formula K : , where and are the similarity data and weight factors corresponding to the feature vector correlation subset respectively, r is the number of feature vector correlation subsets, and w and v are the numbers of the corresponding combined knowledge feature points respectively; Obtain the similarity correlation between the corresponding education knowledge information according to the control similarity data corresponding to the vector correlation control group, and conduct induction and integration on each evaluation knowledge point vector in the initial distribution vector space according to the corresponding similarity correlation to generate an evaluation knowledge point evaluation vector space.

[0009] Furthermore, the process of generating the evaluation knowledge storage space includes: Obtain education evaluation information, disassemble and process the education evaluation information to obtain the characteristic attributes corresponding to the evaluation process and evaluation objectives corresponding to the education evaluation information; Compare and analyze the obtained characteristic attributes with the corresponding evaluation knowledge point vectors in the evaluation knowledge point evaluation vector space to obtain the characteristic knowledge points matching the corresponding education evaluation information; Sort according to the connection relationship of the evaluation knowledge point evaluation vector space to which the characteristic knowledge points corresponding to the education evaluation information belong and the corresponding control similarity data. Set an evaluation retrieval number for the education evaluation information according to the corresponding sorting result, set a number storage subspace at the connection relationship of the corresponding knowledge feature points in the evaluation knowledge point evaluation vector space, and store the corresponding evaluation retrieval number into the corresponding number storage subspace; Set an evaluation storage library, store the corresponding education evaluation information uniformly into the evaluation storage library, and retrieve the corresponding education evaluation information in the evaluation storage library through the number storage subspace according to the corresponding evaluation retrieval number to generate an evaluation knowledge storage space.

[0010] Furthermore, the process of obtaining the personalized evaluation feedback information corresponding to the user includes: Set corresponding initial extraction rules according to the basic education evaluation information, obtain the corresponding education evaluation information in the evaluation knowledge storage space according to the initial extraction plan, and generate initial evaluation information; Obtain user behavior data, correlate the corresponding evaluation answering process data and evaluation answering result data in the user behavior data with the corresponding education evaluation data involved in the initial evaluation information, and set an evaluation control group according to the correlation result; Set the knowledge feature points involved in the education evaluation data in the evaluation control group as the headers corresponding to the corresponding evaluation analysis tables in sequence, set the corresponding fields according to the answering process data, answering time data, and answering detail data corresponding in the evaluation answering process data and the evaluation answering result data, and set the evaluation analysis tables corresponding to the corresponding evaluation control groups; Set the standard evaluation data of the corresponding cells based on the headers and fields in the evaluation analysis table, and obtain the evaluation results of the cells corresponding to the evaluation answering process data in the corresponding number of evaluation analysis tables; Compare and analyze the obtained evaluation results with the evaluation answering result data, form a personalized data set with the corresponding evaluation results and the evaluation answering result data, and analyze and train the personalized data set based on the deep learning algorithm to construct a personalized analysis model; Perform adaptive planning processing on the standard evaluation data and evaluation coefficients of the field cells corresponding to the corresponding user numbers based on the personalized analysis model to obtain the corresponding personalized field coefficients; Analyze and process the corresponding education evaluation data and user behavior data in the initial evaluation information in sequence according to the personalized field coefficients corresponding to the user numbers, obtain the evaluation results of the corresponding knowledge feature points, and generate personalized evaluation feedback information according to the evaluation results of the initial evaluation information.

[0011] Furthermore, the process of obtaining the user's education evaluation results includes: Obtain personalized evaluation feedback information, where the personalized evaluation feedback information includes the personalized field coefficients and evaluation results of the evaluation answering process data corresponding to the corresponding knowledge feature points; Obtain corresponding evaluation feedback requirements according to the evaluation results of corresponding knowledge feature points. The evaluation feedback requirements include multiple related knowledge feature points. Obtain evaluation retrieval characters according to the corresponding knowledge feature points, obtain the evaluated retrieval numbers in the corresponding numbered storage subspaces in the evaluation knowledge storage space according to the evaluation retrieval characters, and obtain corresponding education evaluation information according to the obtained evaluated retrieval numbers; Dynamically feedback the obtained education evaluation information to the user evaluation process, obtain the user behavior data corresponding to the corresponding education evaluation information and the corresponding evaluation results until the corresponding user education evaluation results are obtained.

[0012] The present invention has the following beneficial effects: 0. In the present invention, corresponding basic education evaluation information is submitted according to the evaluation requirements of evaluation staff. Different education evaluation scopes, objectives, and evaluation subjects are obtained through the basic education evaluation information, which improves the personalization in the evaluation process to a certain extent. Retrieve corresponding education knowledge information and education evaluation information according to personalized needs, thereby improving the accuracy of the education auxiliary evaluation results to a certain extent, and avoiding adding a burden to the education evaluation process due to retrieving a large amount of redundant information according to keywords.

[0013] 1. In the present invention, by analyzing and processing education knowledge information, set the knowledge feature points corresponding to the corresponding education knowledge information, and construct a corresponding evaluation knowledge point evaluation vector space. Store the education evaluation information in association through the evaluation knowledge point evaluation vector space, thereby improving the efficiency in the process of obtaining education evaluation information and improving the convenience in the retrieval process.

[0014] 2. In the present invention, judge the personalized data of the corresponding user in the evaluation process through the initial evaluation information for the education evaluation information and the corresponding user behavior data in the corresponding user evaluation process. Dynamically adjust the corresponding evaluation criteria according to the personalized data, reducing the error caused by personalized differences in the evaluation process to a certain extent. Through the feedback of the evaluation results, obtain the corresponding education evaluation information, and conduct multiple evaluations on the user through the secondarily obtained education evaluation information, thereby improving the accuracy of the education evaluation auxiliary system to a certain extent. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic structural diagram of an evaluation auxiliary system based on artificial intelligence proposed by the present invention; Figure 2 is a schematic flow diagram of an evaluation auxiliary method based on artificial intelligence proposed by the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0017] Embodiment 1 As Figure 1 shown, an evaluation assistance system based on artificial intelligence proposed by the present invention includes an education evaluation assistance center, and the education evaluation assistance center includes an intelligent acquisition module, an intelligent analysis module, an intelligent storage module, an adaptive management module, and an evaluation analysis module.

[0018] In this embodiment, the education evaluation assistance center is used to set a personalized evaluation process for the corresponding user according to the learning data in the user's learning process, conduct evaluation feedback on the user's learning process according to the personalized evaluation results of the corresponding user, and provide personalized assistance to the user's learning process according to the evaluation feedback results. The specific implementation process includes: The intelligent acquisition module is used to collect educational knowledge information, educational evaluation information, and user behavior data in the user's learning process of the corresponding educational evaluation content. The specific implementation process includes: Set up an education collection unit and a user collection unit; The education collection unit is used to collect educational knowledge information and educational evaluation information corresponding to the educational evaluation content. The specific implementation process includes: Set up an evaluation management window, which is used for the corresponding educational evaluation staff to enter the basic information of the educational evaluation. The basic information of the educational evaluation includes the corresponding educational evaluation scope information, educational evaluation target information, and educational evaluation informed information, where: The educational evaluation scope information includes evaluation level information, evaluation content information, and evaluation subject information; The educational evaluation target information includes the corresponding evaluation intention and the result information expected to be achieved; The educational evaluation informed information includes the description of the evaluation purpose corresponding to the evaluation subject, the notification of the evaluation content and scope, the introduction of the evaluation methods and means, the information on the use of the evaluation results, the informed consent form of the evaluation subject, etc.; Verify and analyze the obtained basic information of the educational evaluation, analyze and process the basic information of the educational evaluation that passes the verification analysis, obtain the educational evaluation scope information corresponding to the basic information of the educational evaluation, and obtain educational knowledge information and educational evaluation information according to the educational evaluation scope information; Obtain educational evaluation scope information. The evaluation level information includes the vocational types corresponding to vocational education; the evaluation content information includes corresponding content types such as knowledge point information and skill point information; the evaluation subject information includes teachers or students. Set up multiple information retrieval sub-windows according to the content types included in the educational evaluation scope information. The information retrieval sub-windows respectively correspond to the retrieval permissions of the corresponding vocational education types, educational content, and educational resource channels corresponding to the educational subjects. Based on the information retrieval sub-windows, monitor the educational knowledge information and educational evaluation information corresponding to the corresponding educational evaluation scope information in real time through the corresponding retrieval permissions. The educational knowledge information includes the educational planning information and educational knowledge point information of the evaluation content information corresponding to the corresponding educational evaluation scope information. The educational evaluation information is the educational information used for test evaluation during the user evaluation process. The user acquisition unit is used to connect with the corresponding intelligent acquisition device, and collect user behavior data during the educational test evaluation process through the corresponding intelligent acquisition device. Its specific implementation process includes: Obtain the corresponding user basic information during the educational test evaluation process, generate the corresponding user number according to the user basic information, and send the corresponding user number into the corresponding intelligent acquisition device. The corresponding intelligent acquisition device identifies the user according to the corresponding user number, obtains the corresponding identification verification result, and conducts a test evaluation on the user corresponding to the identification verification result. Monitor the test evaluation process of the corresponding user through the intelligent acquisition device. The intelligent acquisition device includes visual sensors, time sensors, motion perception sensors, etc., and collect the corresponding user behavior data through the corresponding intelligent acquisition device. The user behavior data includes the evaluation answering process data and evaluation answering result data of the corresponding user during the educational test evaluation process. The evaluation answering process data includes answering process data, answering time data, and answering detail data. The answering process data includes the corresponding operation steps, and the answering detail data includes limb movement data that has nothing to do with the evaluation result, etc. Temporarily store the obtained educational evaluation information, educational knowledge information, and user behavior data, and send them to other modules.

[0019] The intelligent analysis module is used to systematically analyze the collected educational knowledge information, obtain the corresponding evaluation knowledge point vectors, and conduct a similarity correlation analysis on the obtained evaluation knowledge point vectors to construct an evaluation knowledge point evaluation vector space. Its specific implementation process includes: Set up a knowledge analysis unit and an evaluation analysis unit. The knowledge analysis unit is used to systematically analyze the educational knowledge information corresponding to the corresponding basic educational evaluation information, and obtain the corresponding evaluation knowledge point vectors. The specific implementation process includes: Obtain the corresponding educational planning information and educational knowledge point information in the educational knowledge information, and sort the corresponding educational knowledge point information in sequence according to the educational planning information; set knowledge nodes for the corresponding educational knowledge points according to the sorting process; Based on the knowledge nodes, analyze and process the characteristic attributes corresponding to the corresponding educational knowledge point information. The specific implementation process includes: Extract knowledge characteristics from the knowledge point definition information corresponding to the educational knowledge point information to obtain corresponding knowledge characteristic points, and obtain corresponding association relationships for the obtained knowledge characteristic points based on natural language processing technology. The association relationships include causal relationships, upper and lower position relationships, parallel relationships, etc. Obtain the corresponding relationships between each knowledge characteristic point according to the association analysis results, and set the corresponding relationships as edges; Connect the obtained knowledge characteristic points and corresponding edges to construct a knowledge point characteristic graph structure; Allocate corresponding initial characteristic data to the corresponding knowledge characteristic points according to the knowledge point characteristic graph structure; Analyze and process the knowledge point characteristic graph structure according to the corresponding initial characteristic data, and analyze and train the corresponding knowledge point characteristic graph structure based on the graph neural network algorithm, where: Obtain the characteristic loss function S corresponding to the corresponding knowledge point characteristic graph structure: , M is the number of corresponding edges, is the set of edges, is the true label (0 or 1) indicating whether there is an edge between knowledge characteristic point i and knowledge characteristic point j, is the probability that the model predicts that there is an edge between knowledge characteristic point i and knowledge characteristic point j; Construct a corresponding graph neural network according to the corresponding characteristic loss function S; Until the characteristic loss function S tends to be stable, output the characteristic vectors corresponding to the corresponding knowledge characteristic points in the corresponding educational knowledge point information, obtain the characteristic vector set corresponding to the corresponding educational knowledge point information, perform marking processing according to the characteristic vector set to which the corresponding characteristic vectors belong, and integrate each characteristic vector in the corresponding characteristic vector set to obtain the evaluation knowledge point vector corresponding to the corresponding educational knowledge point information.

[0020] The evaluation analysis unit is used to perform similarity and relevance analysis according to the evaluation knowledge point vectors obtained by the knowledge analysis unit, construct an evaluation knowledge point evaluation vector space, and send the obtained evaluation knowledge point evaluation vector space to the intelligent storage module. The specific implementation process includes: Obtain the education planning information corresponding to the education knowledge information and the evaluation knowledge point vectors corresponding to the education knowledge points information; According to the education planning information, preliminarily arrange and distribute the evaluation knowledge point vectors corresponding to the corresponding education knowledge point information, and construct an initial distribution vector space; Conduct a similarity correlation analysis on the evaluation knowledge point vectors corresponding to the corresponding education knowledge point information in the initial distribution vector space. The specific implementation process includes: Successively set corresponding vector correlation control groups for each evaluation knowledge point vector in the initial distribution vector space. The vector correlation control group includes two evaluation knowledge point vectors; Perform a splitting process on the evaluation knowledge point vectors in the vector correlation control group according to the corresponding integration process to obtain the feature vectors corresponding to each knowledge feature point in the feature vector set; Combine the feature vectors in the corresponding feature vector set in the vector correlation control group to obtain a feature vector correlation subset, and obtain the number R of the corresponding feature vector correlation subsets; Conduct a correlation analysis on the feature vectors in the feature vector correlation subset, mark the corresponding feature vectors as A and B respectively, and mark the similarity data corresponding to the feature vector correlation subset as , where: , integrate the similarity data corresponding to the corresponding feature vector correlation subset, set the importance coefficient corresponding to the corresponding knowledge feature point, and set the weight factor corresponding to the corresponding feature vector correlation subset according to the importance coefficient , where: ; Preset a feature correlation threshold TY, and compare and analyze the similarity data corresponding to the corresponding feature vector correlation subset with the feature correlation threshold: If TY≥ , then there is no corresponding correlation for this feature vector correlation subset; If TY< , then there may be a corresponding correlation for this feature vector correlation subset, and mark the corresponding feature vector correlation subset; Analyze and process the feature vector correlation subsets that may have correlations in the vector correlation control group to obtain the control similarity data of the vector correlation control group, marked as CT K , where K is the number of vector correlation control groups, where: ; Obtain the similarity correlation between the corresponding control similarity data of the vector-associated control group and the corresponding educational knowledge information; re-label each evaluation knowledge point vector in the initial distribution vector space according to the corresponding control similarity data, and process it according to the control similarity based on the spatial position of the evaluation knowledge point vector after the re-labeling process to generate the corresponding evaluation knowledge point evaluation vector space, and send the obtained evaluation knowledge point evaluation vector space to the intelligent storage module.

[0021] The intelligent storage module is used to perform matching analysis on the educational evaluation information and the corresponding educational knowledge information, map the corresponding educational evaluation information into the evaluation knowledge evaluation vector space according to the matching analysis result to generate an evaluation knowledge storage space, and store it. The specific implementation process includes: Obtain the educational evaluation information, disassemble the educational evaluation information, and obtain the characteristic attributes corresponding to the evaluation process and evaluation objectives corresponding to the educational evaluation information; Compare and analyze the obtained characteristic attributes with the corresponding evaluation knowledge point vectors in the evaluation knowledge point evaluation vector space to obtain the characteristic knowledge points that match the corresponding educational evaluation information, and set the characteristic knowledge points corresponding to the corresponding matching results as the evaluation associated knowledge points corresponding to the corresponding educational evaluation information; It should be further noted that in the specific implementation process, the educational evaluation information may include multiple evaluation associated knowledge points or only one evaluation associated knowledge point; The educational evaluation information sets an educational evaluation code according to the evaluation associated knowledge points corresponding thereto. The specific implementation process includes: Obtain the connection relationship of the evaluation associated knowledge points corresponding to the educational evaluation information in the evaluation knowledge point evaluation vector space, and sort the educational evaluation information according to the correlation degree of the control similarity data between the evaluation associated knowledge points; Set a main label code according to each educational knowledge information in the evaluation knowledge point evaluation vector space, and set a corresponding sub-label code for the knowledge characteristic points corresponding to the corresponding educational knowledge information according to the main label code; Obtain the editing codes corresponding to the evaluation associated knowledge points of the educational evaluation information, sort the corresponding label code information according to the sorting result of the evaluation associated knowledge points to generate an evaluation retrieval number; Map the evaluation retrieval number corresponding to the corresponding educational evaluation information to the connection relationship of the corresponding knowledge characteristic points in the evaluation knowledge point evaluation vector space, generate a number storage subspace according to the corresponding connection relationship, and store the corresponding evaluation retrieval number in the corresponding number storage subspace; Set up an evaluation storage repository in the evaluation knowledge storage space, uniformly store the corresponding education evaluation information in the evaluation storage repository, and retrieve the corresponding education evaluation information in the evaluation storage repository through the numbered storage subspace according to the corresponding evaluation retrieval number. Map the numbered storage subspace to the corresponding connection position in the evaluation knowledge point evaluation vector space to generate the evaluation knowledge storage space; It should be further noted that in the specific implementation process, the evaluation knowledge storage space stores the educational knowledge point data corresponding to different professional types of vocational education in an associated manner, thereby improving the association corresponding to different professional types in the process of vocational education, expanding the educational thinking for students or teachers in the process of educational evaluation, and also facilitating the evaluation subject to understand educational knowledge in more fields.

[0022] The adaptive management module is used to obtain the corresponding user behavior data, compare and analyze the user behavior data with the corresponding education evaluation information in the evaluation knowledge storage space, obtain the personalized evaluation feedback information corresponding to the user according to the comparison and analysis results, and send the personalized evaluation feedback information to the evaluation analysis module. Its specific implementation process includes: Set up an evaluation planning unit and an evaluation feedback unit; The evaluation planning unit is used to obtain the basic education evaluation information corresponding to the corresponding user number and the corresponding evaluation knowledge storage space; According to the basic education evaluation information, set the corresponding initial extraction rules, obtain the corresponding education evaluation information in the evaluation knowledge storage space according to the initial extraction plan, generate the initial evaluation information, and send the obtained initial evaluation information to the corresponding user number; It should be further noted that in the specific implementation process, the user behavior data is the monitoring result of the user corresponding to the user number in the process of answering the corresponding initial evaluation information; the initial evaluation information includes a corresponding number of education evaluation information. In the process of obtaining the education evaluation information, it is selected according to the length of the evaluation retrieval number corresponding to the education evaluation information and the knowledge feature points involved. And the initial extraction rules include the number of knowledge feature points involved, the number of other knowledge feature points involved, the difficulty degree of the number of knowledge feature points involved in the education evaluation information, and the quantity plan of the education evaluation information with the corresponding difficulty degree, etc.; The evaluation feedback unit is used to obtain the user behavior data corresponding to the corresponding user number and the corresponding initial evaluation information, analyze and process the user behavior data corresponding to the corresponding education evaluation information in the initial evaluation information, and obtain the personalized evaluation feedback information. Its specific implementation process includes: Correspondingly associate the evaluation answering process data and evaluation answering result data in the user behavior data with the corresponding education evaluation data involved in the initial evaluation information, and set up an evaluation control group according to the association result; Obtain the corresponding knowledge feature points for the corresponding education evaluation data in each evaluation control group according to the evaluation retrieval number, and set up an evaluation analysis table corresponding to the evaluation control group for the knowledge feature points, evaluation answering process data, and evaluation answering result data; Sequentially set the knowledge feature points as the table headers corresponding to the corresponding evaluation analysis tables, and sequentially set the corresponding fields according to the answering process data, answering time data, and answering detail data corresponding in the evaluation answering process data and the evaluation answering result data; Select a corresponding number of evaluation analysis tables, and based on the artificial intelligence algorithm, analyze and process the corresponding evaluation answering process data and evaluation answering result data in the corresponding number of evaluation analysis tables for the data in the fields corresponding to the corresponding knowledge feature points, and set two-way personalized evaluation results respectively according to the evaluation answering process data and evaluation answering result data. The specific implementation process includes: Based on the corresponding table headers and field setting standard evaluation data in the evaluation analysis table, obtain the evaluation results of the corresponding evaluation answering process data cells in the corresponding number of evaluation analysis tables for standard evaluation; Mark the personalized evaluation result data as GX, and mark the answering process data, answering time data, and answering detail data and the corresponding standard evaluation data as G, T, and D respectively, and , mark the corresponding evaluation coefficient as , where: ; Compare and analyze the obtained evaluation results with the evaluation answering result data, form a personalized data set with the corresponding evaluation results and the evaluation answering result data, and analyze and train the personalized data set based on the deep learning algorithm to construct a personalized analysis model; perform adaptive planning processing on the standard evaluation data and evaluation coefficient of the field cells corresponding to the corresponding user numbers based on the personalized analysis model; Obtain the adaptive planning processing results corresponding to the corresponding number of evaluation analysis tables, and obtain the personalized field coefficients of the corresponding user numbers; Evaluate the evaluation results of the corresponding knowledge feature points in the corresponding education evaluation data in each evaluation control group according to the corresponding personalized field coefficients, and feedback the evaluation results to the corresponding cells; Plan and integrate the marked results in the corresponding cells in the evaluation comparison tables corresponding to each evaluation control group obtained, and judge the initial evaluation results of the corresponding knowledge feature points; Analyze and process the corresponding education assessment data and user behavior data in the initial assessment information in sequence according to the personalized field coefficients corresponding to the user numbers, obtain the evaluation results of the corresponding knowledge feature points, and generate personalized assessment feedback information based on the evaluation results of the initial assessment information.

[0023] The said assessment analysis module is used to obtain the associated education assessment information stored in the assessment knowledge storage space according to the personalized assessment feedback information, dynamically feedback the education assessment information to the user assessment process, and obtain the corresponding user's education assessment results. Its specific implementation process includes: Obtain the personalized assessment feedback information, which includes the personalized field coefficients and evaluation results of the assessment answering process data corresponding to the corresponding knowledge feature points; Obtain the corresponding evaluation feedback requirements according to the evaluation results of the corresponding knowledge feature points. The evaluation feedback requirements include multiple associated knowledge feature points. Obtain the assessment retrieval character according to the corresponding knowledge feature points, obtain the assessment retrieval number in the corresponding numbered storage subspace in the assessment knowledge storage space according to the assessment retrieval character, and obtain the corresponding education assessment information according to the obtained assessment retrieval number; Dynamically feedback the obtained education assessment information to the user assessment process, obtain the user behavior data corresponding to the corresponding education assessment information and the corresponding evaluation results until the corresponding user's education assessment results are obtained.

[0024] As Figure 2 shown, an assessment assistance method based on artificial intelligence includes the following steps: Step 1: Collect the education knowledge information, education assessment information for the user education assessment process, and the user behavior data during the user assessment process; Step 2: Conduct a systematic analysis of the collected education knowledge information, obtain the corresponding assessment knowledge point vectors, and construct an assessment knowledge point evaluation vector space according to the assessment knowledge point vectors; Step 3: Perform a matching analysis on the education assessment information and the corresponding education knowledge information, and plan and store the corresponding education assessment information in the assessment knowledge evaluation vector space according to the matching analysis results to generate an assessment knowledge storage space; Step 4: Conduct a personalized evaluation on the user behavior data, the assessment results and assessment processes of the knowledge feature points corresponding to the education assessment information, and obtain the personalized assessment feedback information corresponding to the user; Step 5: Obtain the associated education assessment information stored in the assessment knowledge storage space according to the personalized assessment feedback information, dynamically feedback the education assessment information to the user assessment process, and obtain the corresponding user's education assessment results.

[0025] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An evaluation assistance system based on artificial intelligence, including an education evaluation assistance center, characterized in that, The education evaluation assistance center includes an intelligent acquisition module, an intelligent analysis module, an intelligent storage module, an adaptive management module, and an evaluation analysis module; The intelligent acquisition module is used to collect educational knowledge information, educational evaluation information, and user behavior data during the user evaluation process; The intelligent analysis module is used to systematically analyze the collected educational knowledge information, obtain the corresponding evaluation knowledge point vectors, and construct an evaluation knowledge point evaluation vector space according to the evaluation knowledge point vectors; The intelligent storage module is used to perform matching analysis on the educational evaluation information and the corresponding educational knowledge information, and plan and store the corresponding educational evaluation information in the evaluation knowledge evaluation vector space according to the matching analysis result to generate an evaluation knowledge storage space; The adaptive management module is used to perform personalized evaluation on the evaluation results and evaluation process of the knowledge feature points corresponding to the user behavior data and the educational evaluation information, and obtain the personalized evaluation feedback information corresponding to the user; The evaluation analysis module is used to obtain the relevant educational evaluation information stored in the evaluation knowledge storage space according to the personalized evaluation feedback information, dynamically feedback the educational evaluation information to the user evaluation process, and obtain the educational evaluation results of the corresponding user.

2. The evaluation assistance system based on artificial intelligence according to claim 1, wherein, The process of collecting educational knowledge information, educational evaluation information, and user behavior data during the user evaluation process includes: Set up an evaluation management window, which is used for the corresponding educational evaluation staff to enter the basic information of educational evaluation. The basic information of educational evaluation includes the corresponding educational evaluation scope information, educational evaluation target information, and educational evaluation informed information; set up multiple information retrieval sub-windows according to the educational evaluation scope information, and obtain educational knowledge information and educational evaluation information through the corresponding information retrieval sub-windows; Obtain the corresponding user basic information during the educational test evaluation process, monitor the test evaluation process of the corresponding user through intelligent acquisition devices, and obtain the corresponding user behavior data. The user behavior data includes evaluation answering process data and evaluation answering result data.

3. The evaluation assistance system based on artificial intelligence according to claim 2, wherein, The process of obtaining the evaluation knowledge point vectors includes: Obtain educational knowledge information, which includes educational planning information and educational knowledge point information. Extract the knowledge feature of the corresponding knowledge point definition information in the educational knowledge point information to obtain the corresponding knowledge feature points; Obtain the association relationship between the corresponding knowledge feature points based on natural language processing technology, and construct a knowledge point feature map structure according to the corresponding knowledge feature points and association relationship; Analyze and train the corresponding knowledge point feature map structure based on the graph neural network algorithm, output the feature vectors corresponding to each knowledge feature point in the corresponding educational knowledge point information, obtain the feature vector set corresponding to the corresponding educational knowledge point information, and integrate and process each feature vector in the corresponding feature vector set to obtain the evaluation knowledge point vector corresponding to the corresponding educational knowledge point information.

4. The evaluation assistance system based on artificial intelligence according to claim 3, wherein The process of constructing the evaluation knowledge point evaluation vector space includes: Perform preliminary arrangement and distribution on the evaluation knowledge point vectors corresponding to the corresponding educational knowledge point information according to the educational planning information to construct an initial distribution vector space; Perform a similarity correlation analysis on the evaluation knowledge point vectors corresponding to each educational knowledge point information in the initial distribution vector space, sequentially set corresponding vector correlation control groups for each evaluation knowledge point vector, split the evaluation knowledge point vectors in the vector correlation control groups according to the corresponding integration process, and obtain the feature vectors corresponding to each knowledge feature point in the feature vector set; Obtain a feature vector correlation subset according to the corresponding feature vectors, perform a correlation analysis on the feature vectors in the feature vector correlation subset, and determine whether there may be a correlation in the corresponding feature vector correlation subset; Analyze and process the eigenvector correlation subset that may be correlated within the vector correlation control group, and obtain the control similarity data CT of the corresponding vector correlation control group through the following formula K : , where and are the similarity data and weight factors corresponding to the feature vector correlation subsets respectively, r is the number of feature vector correlation subsets, and w and v are the numbers of the corresponding combined knowledge feature points respectively; Obtain the similarity correlation between the corresponding educational knowledge information according to the control similarity data corresponding to the vector correlation control group, and perform an inductive integration on each evaluation knowledge point vector in the initial distribution vector space according to the corresponding similarity correlation to generate an evaluation knowledge point evaluation vector space.

5. The evaluation assistance system based on artificial intelligence according to claim 4, characterized in that The process of generating the evaluation knowledge storage space includes: Obtain educational evaluation information, perform a decomposition process on the educational evaluation information, and obtain the characteristic attributes corresponding to the evaluation process and evaluation objectives corresponding to the educational evaluation information; Compare and analyze the obtained characteristic attributes with the corresponding evaluation knowledge point vectors in the evaluation knowledge point evaluation vector space to obtain the characteristic knowledge points that match the corresponding educational evaluation information; Sort according to the connection relationship of the evaluation knowledge point evaluation vector space to which the characteristic knowledge points corresponding to the educational evaluation information belong and the corresponding control similarity data; Set an evaluation retrieval number for the educational evaluation information according to the corresponding sorting result, set a numbered storage subspace at the connection relationship of the corresponding knowledge feature points in the evaluation knowledge point evaluation vector space, and store the corresponding evaluation retrieval number in the corresponding numbered storage subspace; Set an evaluation storage library, uniformly store the corresponding educational evaluation information in the evaluation storage library, and retrieve the corresponding educational evaluation information in the evaluation storage library through the numbered storage subspace according to the corresponding evaluation retrieval number to generate an evaluation knowledge storage space.

6. The evaluation assistance system based on artificial intelligence according to claim 5, characterized in that, The process of obtaining the personalized evaluation feedback information corresponding to the user includes: Set corresponding initial extraction rules according to the basic educational evaluation information, obtain the corresponding educational evaluation information in the evaluation knowledge storage space according to the initial extraction plan, and generate initial evaluation information; Obtain user behavior data, perform a corresponding association between the corresponding evaluation answering process data and evaluation answering result data in the user behavior data and the corresponding educational evaluation data involved in the initial evaluation information, and set an evaluation control group according to the association result; Sequentially set the knowledge feature points involved in the educational evaluation data in the evaluation control group as the headers corresponding to the corresponding evaluation analysis tables, sequentially set the corresponding fields according to the answering process data, answering time data, and answering detail data corresponding to the answering process data in the evaluation answering process data and the evaluation answering result data, and set the evaluation analysis tables corresponding to the corresponding evaluation control groups; Set the standard evaluation data for the corresponding cells based on the headers and fields in the evaluation analysis table, and obtain the evaluation results of the cells corresponding to the evaluation answering process data in the corresponding number of evaluation analysis tables; Compare and analyze the obtained evaluation results with the test answer result data, form a personalized data set with the corresponding evaluation results and test answer result data, and analyze and train the personalized data set based on the deep learning algorithm to construct a personalized analysis model; Based on the personalized analysis model, perform adaptive planning processing on the standard evaluation data and evaluation coefficients of the field cells corresponding to the corresponding user ID to obtain the corresponding personalized field coefficients; Analyze and process the corresponding education evaluation data and user behavior data in the initial evaluation information in turn according to the personalized field coefficients corresponding to the user ID, obtain the evaluation results of the corresponding knowledge feature points, and generate personalized evaluation feedback information according to the evaluation results of the initial evaluation information.

7. An evaluation assistance system based on artificial intelligence according to claim 6, characterized in that, The process of obtaining the user's education evaluation results includes: Obtain personalized evaluation feedback information, which includes the personalized field coefficients and evaluation results of the test answer process data corresponding to the corresponding knowledge feature points; Obtain the corresponding evaluation feedback requirements according to the evaluation results of the corresponding knowledge feature points. The evaluation feedback requirements include multiple related knowledge feature points. Obtain the test retrieval characters according to the corresponding knowledge feature points, obtain the test retrieval numbers in the corresponding numbered storage subspace in the test knowledge storage space according to the test retrieval characters, and obtain the corresponding education evaluation information according to the obtained test retrieval numbers; Dynamically feedback the obtained education evaluation information to the user's evaluation process, obtain the user behavior data corresponding to the corresponding education evaluation information and the corresponding evaluation results until the corresponding user's education evaluation results are obtained.

8. A method for assisting in evaluation based on artificial intelligence corresponding to an evaluation assistance system based on artificial intelligence according to any one of claims 1 to 7, characterized in that, It includes the following steps: Step 1: Collect the education knowledge information, education evaluation information for the user's education evaluation process, and the user behavior data during the user's evaluation process; Step 2: Conduct a systematic analysis of the collected education knowledge information, obtain the corresponding evaluation knowledge point vectors, and construct an evaluation knowledge point evaluation vector space according to the evaluation knowledge point vectors; Step 3: Perform a matching analysis on the education evaluation information and the corresponding education knowledge information, and plan and store the corresponding education evaluation information in the evaluation knowledge evaluation vector space according to the matching analysis results to generate an evaluation knowledge storage space; Step 4: Perform a personalized evaluation on the user behavior data, the evaluation results and evaluation processes of the knowledge feature points corresponding to the education evaluation information to obtain the personalized evaluation feedback information corresponding to the user; Step 5: Obtain the related education evaluation information stored in the evaluation knowledge storage space according to the personalized evaluation feedback information, dynamically feedback the education evaluation information to the user's evaluation process, and obtain the education evaluation results of the corresponding user.

Images