AI automatic reviewing system based on personalized learning model

Abstract

The invention relates to the technical field of artificial intelligence, in particular to an AI automatic reviewing system based on a personalized learning model, which comprises an expert knowledge base; the evaluation model is constructed based on historical academic conditions and historical activity conditions of historical students and corresponding historical comment results, and is used for providing a first target comment result for a target experiment report of a target student; the personalized review module is connected with the expert knowledge base and the evaluation model, and is used for calling the content of the expert knowledge base based on the expert knowledge base and the evaluation model and providing a review result for the target experiment report when the target experiment report uploaded by the target student is received; the feedback module is used for receiving first feedback information of the target student and second feedback information of the teacher; and the first optimization module is connected with the feedback module and the personalized review module, and is used for optimizing the personalized review module based on the first feedback information and the second feedback information so as to obtain a personalized review conclusion better fitting personal learning characteristics.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an AI automatic review system based on a personalized learning model. Background Art

[0002] Traditional automatic review systems usually use uniform standards to evaluate each student's lab report, ignoring individual differences and personalized needs of students. However, with the popularization of personalized learning concepts, educators and students can expect to obtain review feedback that is more in line with personal learning characteristics.

[0003] Therefore, how to obtain personalized review conclusions that are more in line with individual learning characteristics is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of the above problems, the present invention provides an AI automatic review system based on a personalized learning model that overcomes the above problems or at least partially solves the above problems.

[0005] The present invention provides an AI automatic review system based on a personalized learning model, comprising:

[0006] An expert knowledge base storing first information related to the experimental content, second information related to the opinions and comments of the experts, common errors, evaluation criteria, and corresponding improvement suggestions;

[0007] The evaluation model is constructed based on the history students’ history academic situation, history activity situation and corresponding history review results, and is used to provide the first target review results for the target students’ target experimental reports;

[0008] A personalized review module, connected to the expert knowledge base and the evaluation model, for invoking the content of the expert knowledge base based on the expert knowledge base and the evaluation model upon receiving the target experiment report uploaded by the target student, and providing a review result for the target experiment report;

[0009] A feedback module, used for receiving first feedback information from a target student and second feedback information from a target teacher;

[0010] The first optimization module is connected to the feedback module and the personalized review module, and is used to optimize the personalized review module based on the first feedback information and the second feedback information.

[0011] Preferably, the first information includes: textbooks, academic papers and experimental manuals in various subject areas;

[0012] The evaluation criteria include: standards and indicators for the quality of experimental reports.

[0013] Preferably, the construction process of the evaluation model is as follows:

[0014] Determine the abstract ability map of each history student based on the history academic performance data, history interactive activity data and corresponding history review results of the history students;

[0015] An evaluation model is trained based on each history student's abstract ability map, history experiment report and corresponding history evaluation results.

[0016] Preferably, the abstract ability map of each history student includes:

[0017] Comprehension ability, analytical ability, collaboration ability and innovation ability as well as the scores for each ability.

[0018] Preferably, it also includes: a personalized recommendation module, connected to the evaluation model, for:

[0019] Obtaining an abstract ability map of the target student;

[0020] Based on the abstract ability map of the target students, personalized learning resources are recommended.

[0021] Preferably, the personalized review module is used to:

[0022] Upon receiving the target experiment report of the target student, extracting key information based on the first target review result;

[0023] Matching the key information with the first information and common errors in the expert knowledge base to obtain a matching result;

[0024] Determining a second target review result based on the matching result, the second information in the expert knowledge base, the evaluation criteria, and the corresponding improvement suggestions;

[0025] Based on the second target review result, a review result is provided for the target experiment report.

[0026] Preferably, the first optimization module includes:

[0027] A judging unit, configured to judge whether the first feedback information and the second feedback information are consistent;

[0028] The first optimization unit is used to adjust the evaluation criteria, common errors and corresponding improvement suggestions in the expert knowledge base and the data set used by the evaluation model through the personalized review module when the judgment unit determines that the result is yes, so as to optimize the personalized review module.

[0029] Preferably, the first optimization module further includes:

[0030] The second optimization unit is used to update the content in the expert knowledge base through the personalized review module at preset intervals.

[0031] Preferably, it also includes:

[0032] The second optimization module connects the personalized recommendation module and the feedback module, and is used to optimize the personalized recommendation module based on the third feedback information of the students and the fourth feedback information of the teachers, wherein the third feedback information and the fourth feedback information are proposed for the personalized learning resources.

[0033] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages:

[0034] The present invention provides an AI automatic review system based on a personalized learning model, comprising: an expert knowledge base storing first information related to experimental content, second information related to expert opinions and comments, common errors, evaluation criteria and corresponding improvement suggestions; an evaluation model constructed based on the historical academic situation, historical activity situation and corresponding historical review results of history students, and used to provide a first target review result for a target experimental report of a target student; a personalized review module connected to the expert knowledge base and the evaluation model, and when receiving the target experimental report uploaded by the target student, used to call the content of the expert knowledge base based on the expert knowledge base and the evaluation model, and provide a review result for the target experimental report of the target student; a feedback module, used to receive the first feedback information of the target student and the second feedback information of the teacher; a first optimization module connected to the feedback module and the personalized review module, and used to optimize the personalized review module based on the first feedback information and the second feedback information, and not only use the evaluation model but also combine with the expert knowledge base to realize personalized review of the student's experimental report, so as to obtain a personalized review conclusion that is more in line with the individual learning characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Also, throughout the accompanying drawings, the same reference figures are used to represent the same components. In the drawings:

[0036] Figure 1 A schematic diagram of the structure of an AI automatic review system based on a personalized learning model in an embodiment of the present invention is shown;

[0037] Figure 2 A schematic diagram of a personalized recommendation module in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0038] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0039] The embodiment of the present invention provides an AI automatic review system based on a personalized learning model, such as Figure 1 As shown, including:

[0040] The expert knowledge base 101 stores first information related to the experimental content, second information related to the expert's opinions and comments, common errors, evaluation criteria, and corresponding improvement suggestions;

[0041] An evaluation model 102 is constructed based on the historical academic situation, historical activity situation and corresponding historical review results of historical academics, and is used to provide a first target review result for the target experimental report of the target student;

[0042] The personalized review module 103 is connected to the expert knowledge base 101 and the evaluation model 102, and is used to call the content of the expert knowledge base 101 based on the expert knowledge base 101 and the evaluation model 102 when receiving the target experiment report uploaded by the target student, and provide a review result for the target experiment report;

[0043] Feedback module 104, used for receiving first feedback information of target students and second feedback information of target teachers;

[0044] The first optimization module 105 is connected to the feedback module 104 and the personalized review module 103 , and is used to optimize the personalized review module 103 based on the first feedback information and the second feedback information.

[0045] First, the expert knowledge base 101 is used to provide the knowledge and review criteria of experts in various fields. Specifically, the expert knowledge base 101 obtains first information related to the experimental content, second information related to the expert's opinions and comments, common errors, evaluation criteria and corresponding improvement suggestions through classification and induction.

[0046] The first information includes textbooks, academic papers and experimental manuals in various disciplines. Specifically, it involves defining the basic scientific concepts and terms that may be used in the experimental report, the steps, methods and expected results of different experiments. For example, for biological experiments, the experimental report must include the content structure, experimental purpose, experimental materials, experimental steps and experimental results.

[0047] The second information related to the expert's opinion comments includes comments on the purpose and background of the experiment from the expert's perspective, comments on the experimental results and data analysis, and comments on the experimental summary and thinking.

[0048] Specifically, comment on whether the purpose of the experiment is clear and reasonable, and whether the theoretical background involved in the experiment is fully understood; evaluate whether the experimental design is reasonable and feasible, whether the experimental steps are clear, and whether the operations are standardized; evaluate whether the experimental results are consistent with expectations, whether the data processing methods are appropriate, and the reliability and rationality of the data analysis results; comment on the comprehensiveness of the experimental summary, the depth of thinking and effectiveness, and whether appropriate reflection and improvement suggestions have been made on the problems in the experiment.

[0049] Common Mistakes, lists common mistakes students may make in their lab reports and how to correct them.

[0050] Evaluation criteria include: standards and indicators of experimental reporting treatment.

[0051] The corresponding improvement suggestions include: improvement suggestions for common errors and problems in the report pointed out by the evaluation criteria.

[0052] The above data are sorted and summarized according to corresponding categories and topics, and the sorted knowledge points are digitized and stored in the expert knowledge base 101, thereby forming the expert knowledge base 101.

[0053] The evaluation model 102 is specifically constructed based on the history students' historical academic situation, historical activity situation and corresponding historical review results, and is used to provide a first target review result for the target student's target experiment report.

[0054] The construction process of the evaluation model 102 is as follows:

[0055] Based on the history students’ historical academic performance data, historical interactive activity data and corresponding historical review results, determine the abstract ability map of each history student;

[0056] An evaluation model is trained based on each history student's abstract ability map, history experiment report and corresponding history evaluation results.

[0057] Among them, the history academic performance data include: data on history students' daily homework, exams and lab reports; the history interactive activity data include: data on history students' classroom interactions, discussions and group projects; the corresponding history review results include: historical review results extracted from literature, overviews and theoretical frameworks.

[0058] The above historical academic performance data, historical interactive activity data and corresponding historical commentary results are processed to obtain the ability dimension division, including the scores of each ability in understanding ability, analysis ability, cooperation ability and innovation ability, thereby determining the abstract ability map of each history student. The data processing process here is an application process of an ability identification model, which is also trained based on a large amount of historical data and is not limited here.

[0059] Next, based on each history student's abstract ability map, history experiment report and corresponding history evaluation results, an evaluation model 102 is trained.

[0060] Specifically, the abstract ability map and history experiment report corresponding to each history student are used as input data, and the history evaluation results are used as output data. Both the input data and the output data are input into the model for training. After repeated iterations, an evaluation model 102 is finally obtained. The evaluation model 102 is used to provide a first target comment result for the target student's target experiment report. The first target comment result combines the personal characteristics of the target student, wherein the personal characteristics are the abstract ability map, including comprehension ability, analysis ability, cooperation ability, innovation ability and the scoring of each ability.

[0061] In actual applications, when the target experiment report of the target student is received, the target historical academic performance data, target historical interactive activity data and corresponding target historical comment results of all target students are extracted based on the target student's personal information; then, based on the target historical academic performance data, target historical interactive activity data and corresponding target historical comment results, the target student's target abstract ability map is determined; then, the target student's target abstract ability map and the target experiment report are input into the evaluation model, thereby obtaining the first target comment result of the target experiment report.

[0062] like Figure 2 As shown, the system further includes: a personalized recommendation module 106, connected to the evaluation model 102, for obtaining the target student's abstract ability map, and recommending personalized learning resources based on the target student's abstract ability map. The personalized learning resources can adapt to the student's learning characteristics.

[0063] By obtaining the abstract ability map of any student, that is, the portrait of any student, and based on the portrait of any student, the personalized recommendation module 106 recommends personalized learning resources suitable for the portrait of the student. The personalized learning resources can take into account the student's interests and learning style.

[0064] Next, the personalized review module 103 connects the expert knowledge base 101 and the evaluation model 102. When receiving the target experiment report uploaded by the target student, it is used to call the content of the expert knowledge base 101 based on the expert knowledge base 101 and the evaluation model 102 to provide a review result for the target experiment report.

[0065] The personalized review module 103 can provide a more professional and targeted review result based on the evaluation model 102 and the content of the expert knowledge base 101 .

[0066] Specifically, the personalized review module 103 is used to:

[0067] When receiving the target experiment report of the target student, extract key information based on the first target review results;

[0068] Matching the key information with the first information and common errors in the expert knowledge base to obtain a matching result;

[0069] Determine a second target review result based on the matching result, the second information in the expert knowledge base, the evaluation criteria, and the corresponding improvement suggestions;

[0070] Based on the second objective review results, provide review results for the lab report.

[0071] On the one hand, the evaluation model 102 can provide the target student's target experiment report with a first target review result, which is only a comment on the ability and is highly personalized. On the other hand, the content in the expert knowledge base 101 can provide comments on professional directions. Therefore, key information is extracted from the first target review result, which can be the title, purpose, etc. of the experimental project, and is matched with the first information and common errors in the expert knowledge base to obtain a matching result, that is, to find the corresponding professional content from the expert knowledge base. Next, the second information, evaluation standards and corresponding improvement suggestions in the expert knowledge base 101 are combined with the above matching results to obtain a professional comment on the experimental content, which is the second target review result. The second target review result adds an evaluation of professional content on the basis of the first target review result, thereby providing a review result for the target experiment report.

[0072] The review results not only include the personalized characteristics of the target students but also professional comments, which can adapt to the development of students with different personalities.

[0073] In addition, the system also includes a feedback module 104 for receiving first feedback information from the target student and second feedback information from the target teacher.

[0074] The system is not only connected to the student client, but also connected to the teacher client, and can receive the first feedback information of the students and the second feedback information of the teachers.

[0075] The first feedback information may be questions or improvements raised by students based on the review results, and the second feedback information may be evaluations or questions made by teachers based on the first feedback information.

[0076] Finally, it also includes: a first optimization module 105, which connects the feedback module 104 and the personalized review module 103, and is used to optimize the personalized review module 103 based on the first feedback information and the second feedback information.

[0077] Specifically, the first optimization module 105 includes:

[0078] A judging unit, configured to judge whether the first feedback information and the second feedback information are consistent;

[0079] The first optimization unit is used to adjust the evaluation criteria, common errors and corresponding improvement suggestions in the expert knowledge base 101 and the data set used in the evaluation model 102 through the personalized review module 103 when the judgment unit determines that the result is yes, so as to optimize the personalized review module 103.

[0080] Among them, the judgment unit is used to judge whether the student's feedback willingness is consistent with the teacher's feedback willingness. Only when they are consistent, the subsequent optimization operation is carried out.

[0081] When judging whether the intentions of the two are the same, a semantic analysis method is used to judge the directionality and degree of judgment of the first feedback information and the second feedback information.

[0082] The optimization of the personalized review module 103 specifically involves optimizing the content in the expert knowledge base 101 through the personalized review module 103, which may specifically include adding, deleting or adjusting corresponding content, and optimizing the data set of the evaluation model 102, which also includes adding, deleting or adjusting corresponding content.

[0083] The first optimization module 105 further includes: a second optimization unit, which is used to update the content in the expert knowledge base 101 through the personalized review module 103 at preset intervals.

[0084] The second optimization unit specifically updates the content in the expert knowledge base 101 according to the latest scientific discoveries and educational trends, so that the expert knowledge base 101 has the most cutting-edge experimental project content.

[0085] In an optional embodiment, if Figure 2As shown, the system also includes: a second optimization module 107, which connects the feedback module 104 and the personalized recommendation module 106, and is used to optimize the personalized recommendation module 106 based on the third feedback information of the target students and the fourth feedback information of the target teachers, and the third feedback information and the fourth feedback information are proposed for personalized learning resources.

[0086] By optimizing the personalized recommendation module 106, the learning resources it provides can be more tailored to individual development.

[0087] Each module in the system can be maintained to ensure its normal operation, and a confidentiality mechanism is added to ensure that student and teacher information will not be leaked.

[0088] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages:

[0089] The present invention provides an AI automatic review system based on a personalized learning model, comprising: an expert knowledge base storing first information related to experimental content, second information related to expert opinions and comments, common errors, evaluation criteria and corresponding improvement suggestions; an evaluation model constructed based on the historical academic situation, activity situation and corresponding historical review results of history students, and used to provide a first target review result for a target experimental report of a target student; a personalized review module connected to the expert knowledge base and the evaluation model, and when receiving the target experimental report uploaded by the target student, used to call the content of the expert knowledge base based on the expert knowledge base and the evaluation model to provide a review result for the target experimental report; a feedback module for receiving first feedback information of the target student and second feedback information of the target teacher; a first optimization module connected to the feedback module and the personalized review module, and used to optimize the personalized review module based on the first feedback information and the second feedback information, so as to realize personalized review of the student's experimental report by not only using the evaluation model but also combining with the expert knowledge base, so as to obtain a personalized review conclusion that is more in line with the individual learning characteristics.

[0090] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0091] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. An AI automatic review system based on a personalized learning model, characterized in that: include: An expert knowledge base storing first information related to the experimental content, second information related to the opinions and comments of the experts, common errors, evaluation criteria, and corresponding improvement suggestions; The evaluation model is constructed based on the history students’ history academic situation, history activity situation and corresponding history review results, and is used to provide the first target review results for the target students’ target experimental reports; A personalized review module, connected to the expert knowledge base and the evaluation model, for invoking the content of the expert knowledge base based on the expert knowledge base and the evaluation model upon receiving the target experiment report uploaded by the target student, and providing a review result for the target experiment report; A feedback module, used for receiving first feedback information from a target student and second feedback information from a target teacher; The first optimization module is connected to the feedback module and the personalized review module, and is used to optimize the personalized review module based on the first feedback information and the second feedback information.

2. The system according to claim 1, characterized in that The first information includes: textbooks, academic papers and experimental manuals in various subject areas; The evaluation criteria include: standards and indicators for the quality of experimental reports.

3. The system according to claim 1, characterized in that The construction process of the evaluation model is as follows: Determine the abstract ability map of each history student based on the history academic performance data, history interactive activity data and corresponding history review results of the history students; An evaluation model is trained based on each history student's abstract ability map, history experiment report and corresponding history evaluation results.

4. The system according to claim 3, characterized in that The abstract ability map of each history student includes: Comprehension ability, analytical ability, collaboration ability and innovation ability as well as the scores for each ability.

5. The system according to claim 3, characterized in that Also includes: The personalized recommendation module is connected to the evaluation model and is used to: Obtaining an abstract ability map of the target student; Based on the abstract ability map of the target students, personalized learning resources are recommended.

6. The system according to claim 3, characterized in that The personalized review module is used to: Upon receiving the target experiment report of the target student, extracting key information based on the first target review result; Matching the key information with the first information and common errors in the expert knowledge base to obtain a matching result; Determining a second target review result based on the matching result, the second information in the expert knowledge base, the evaluation criteria, and the corresponding improvement suggestions; Based on the second target review result, a review result is provided for the target experiment report.

7. The system according to claim 1, characterized in that The first optimization module comprises: A judging unit, configured to judge whether the first feedback information and the second feedback information are consistent; The first optimization unit is used to adjust the evaluation criteria, common errors and corresponding improvement suggestions in the expert knowledge base and the data set used by the evaluation model through the personalized review module when the judgment unit determines that the result is yes, so as to optimize the personalized review module.

8. The system according to claim 7, characterized in that The first optimization module further includes: The second optimization unit is used to update the content in the expert knowledge base through the personalized review module at preset intervals.

9. The system according to claim 5, characterized in that Also includes: The second optimization module connects the personalized recommendation module and the feedback module, and is used to optimize the personalized recommendation module based on the third feedback information of the students and the fourth feedback information of the teachers, wherein the third feedback information and the fourth feedback information are proposed for the personalized learning resources.

Images