Learning condition analysis method, related device, equipment and storage medium

Abstract

The invention discloses a learning condition analysis method, a related device, equipment and a storage medium, and the method comprises the steps: obtaining an evaluation result of a tested object and a difficulty score of a target question; wherein the evaluation result at least comprises a grade score representing the learning level of the tested object; and based on the evaluation result and the difficulty score, generating a learning condition analysis result of the tested object. According to the scheme, the accuracy of learning condition analysis on the tested object can be improved.

Description

[0001] This application claims the priority of the Chinese patent application filed with the State Intellectual Property Office on December 19, 2023, with application number 2023117594646 and application name “Learning Situation Analysis Method and Related Devices, Equipment and Storage Medium”, all contents of which are incorporated by reference in this application. Technical Field

[0002] The present application relates to the field of intelligent education technology, and in particular to a learning situation analysis method and related devices, equipment and storage media. Background Art

[0003] With the continuous development of intelligent education, student evaluation, as an important part of education evaluation, can play a positive guiding, motivating and regulating role in improving students' academic performance and comprehensive quality.

[0004] In the prior art, the learning ability of the subject is usually evaluated based on the average correct rate of the subject's recent answer records. However, since the difficulty of the questions chosen by the subjects varies and the coverage of the answer records is relatively one-sided, the accuracy of the subject's learning situation analysis is low. In view of this, how to improve the accuracy of the subject's learning situation analysis has become an urgent problem to be solved. Summary of the invention

[0005] The main technical problem solved by the present application is to provide a learning situation analysis method and related devices, equipment and storage media, which can improve the accuracy of learning situation analysis of subjects.

[0006] In order to solve the above-mentioned technical problems, the first aspect of the present application provides a learning situation analysis method, including: obtaining the evaluation results of the subject and the difficulty score of the target question; wherein the evaluation results at least include a grade score representing the learning level of the subject; based on the evaluation results and the difficulty score, generating the learning situation analysis result of the subject.

[0007] In order to solve the above-mentioned technical problems, the second aspect of the present application provides a learning situation analysis device, including: an acquisition module and a generation module, the acquisition module is used to obtain the evaluation results of the subject and the difficulty score of the target question; wherein the evaluation results at least include a grade score representing the learning level of the subject; the generation module is used to generate the learning situation analysis results of the subject based on the evaluation results and the difficulty score.

[0008] In order to solve the above-mentioned technical problems, the third aspect of the present application provides an electronic device, comprising a memory and a processor coupled to each other, wherein program instructions are stored in the memory, and the processor is used to execute the program instructions to implement the learning situation analysis method in the above-mentioned first aspect.

[0009] In order to solve the above-mentioned technical problems, the fourth aspect of the present application provides a computer-readable storage medium, which stores program instructions that can be executed by a processor, and the program instructions are used to implement the learning situation analysis method of the above-mentioned first aspect.

[0010] The above scheme generates the learning situation analysis results of the subjects based on the assessment results of the subjects and the difficulty scores of the target questions. The assessment results that at least represent the learning level of the subjects and the difficulty scores that represent the difficulty of the target questions are used as reference information for generating the learning situation analysis results. The learning situation analysis can be performed based on multi-angle data, thereby improving the accuracy of the learning situation analysis of the subjects. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flow chart of an embodiment of the learning situation analysis method of the present application; Figure 2 It is a flow chart of another embodiment of the learning situation analysis method of the present application; Figure 3 It is a schematic diagram of a framework of an embodiment of a model for generating learning situation comments in the learning situation analysis method of the present application; Figure 4 It is a flowchart of another embodiment of the learning situation analysis method of the present application; Figure 5 It is a schematic diagram of an embodiment of the target level distribution of subjects in the learning situation analysis method of the present application; Figure 6 It is a schematic diagram of the framework of an embodiment of the learning situation analysis device of the present application; Figure 7 It is a schematic diagram of the framework of an embodiment of the electronic device of the present application; Figure 8 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0012] The scheme of the embodiment of the present application is described in detail below in conjunction with the drawings of the specification.

[0013] In the following description, for the purpose of explanation rather than limitation, specific details such as specific system structures, interfaces, and technologies are provided to facilitate a thorough understanding of the present application.

[0014] The terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the fragment " / " in this article generally indicates that the associated objects before and after are in an "or" relationship. In addition, "many" in this article means two or more than two.

[0015] See also Figure 1 , Figure 1 This is a flow chart of an embodiment of the learning situation analysis method of the present application. Specifically, it may include the following steps: Step S10: Obtain the test results of the subject and the difficulty score of the target question.

[0016] In the disclosed embodiment, the evaluation result includes at least a grade point representing the learning level of the subject.

[0017] In one implementation scenario, the grade points represent the learning ability of the subject. Specifically, the higher the grade points, the better the learning ability of the corresponding subject, and the lower the grade points, the worse the learning ability of the subject. It should be noted that the upper and lower limits of the grade points of the subjects are not limited in this application.

[0018] In one implementation scenario, when a subject in a preset period first answers a target question, an initial grade is set. It can be understood that the initial grade can be set based on a person skilled in the art, or dynamically set based on the average grade of all subjects in the previous preset period, or adjusted based on the final grade at the end of the previous preset period. Examples are not given one by one here.

[0019] In a specific implementation scenario, the preset period can be determined based on the learning objectives of the subject. For example, the preset period is one semester, or the preset period is three months. Examples are not given one by one here.

[0020] Step S20: Generate the learning situation analysis results of the subject based on the assessment results and difficulty score.

[0021] In the disclosed embodiments, the learning situation analysis results are used to characterize the learning level and problem-solving level of the subjects. Specifically, the learning situation analysis results are presented in text form, numerical form, etc., which is not limited in this application.

[0022] In an implementation scenario, as a possible implementation method, a learning situation analysis model can be pre-trained, and the learning situation analysis model can include but is not limited to a network model of an Encoder-Decoder architecture, etc., and the evaluation results and difficulty scores are input into the learning situation analysis model, and the output results of the learning situation analysis model are used as the learning situation analysis results of the subject. In order to ensure the prediction accuracy of the learning situation analysis model as much as possible, sample data pairs of several sample subjects can be collected, and each group of sample data pairs includes sample evaluation results and sample difficulty scores, as well as the real analysis results of the sample labeled in the sample data pairs. On this basis, the sample data pairs can be processed based on the learning situation analysis model to obtain the sample prediction analysis results of the sample subjects, so that the network parameters of the learning situation analysis model can be adjusted based on the difference between the real analysis results of the samples and the predicted analysis results of the samples, until the training of the learning situation analysis model converges, and the evaluation results of the subjects and the difficulty scores of the target questions can be processed based on the converged learning situation analysis model to obtain the learning situation analysis results of the subjects. It should be noted that the specific processing process of the learning situation analysis model can refer to the technical details such as the network model of the Encoder-Decoder architecture, which will not be repeated here.

[0023] The above scheme generates the learning situation analysis results of the subjects based on the assessment results of the subjects and the difficulty scores of the target questions. The assessment results that at least represent the learning level of the subjects and the difficulty scores that represent the difficulty of the target questions are used as reference information for generating the learning situation analysis results. The learning situation analysis can be performed based on multi-angle data, thereby improving the accuracy of the learning situation analysis of the subjects.

[0024] See also Figure 2 , Figure 2 This is a flow chart of another embodiment of the learning situation analysis method of the present application. Specifically, it may include the following steps: Step S11: Obtain the test results of the subject and the difficulty score of the target question.

[0025] In the disclosed embodiments, the learning situation analysis results include at least learning situation comments about the subjects. Specifically, the learning situation comments are presented in natural language. Natural language is based on the natural, informal language system used by humans and can be used for communication, expression of opinions, and transmission of information. It should be noted that the language type of the learning situation comments is not limited in this application, such as Chinese, English, French, etc.

[0026] In an implementation scenario, the target knowledge point contained in the target question is obtained, and the historical learning data and historical problem-solving data of the subject are obtained. Based on the historical learning data, the historical learning progress of the subject regarding the target knowledge point is determined, and based on the historical problem-solving data, the historical problem-solving accuracy of the subject regarding the target knowledge point is determined. For example, the target question is "It is known that a, b, and c are three consecutive columns of an arithmetic sequence, and a+b+c=21, then what is the value of b", and the knowledge point contained therein is "arithmetic sequence", and the historical learning progress of the subject regarding the target knowledge point can be determined based on the subject's learning progress on courses related to "arithmetic sequence", for example. There are four courses on "arithmetic progression". The subjects have completed the first two courses. It can be obtained that the historical learning progress of the subjects on the target knowledge point is 0.5. Based on the historical learning progress and the historical question-answering accuracy, the subject's grade points are generated, and the historical learning progress is proportional to the grade points, and the historical question-answering accuracy is also proportional to the grade points. It can be understood that the subject's learning progress on the target knowledge point represents the higher the completion of the knowledge point learning, the higher the subject's learning level, and the higher the subject's question-answering accuracy on the target knowledge point, which also represents the higher the subject's learning level. The above scheme, based on the subject's historical learning data and historical question-answering data, obtains a more targeted historical learning progress and historical question-answering accuracy for the target knowledge point, and generates the subject's grade points on the target knowledge point, so it can improve the accuracy of the subject's learning situation analysis.

[0027] In a specific implementation scenario, the historical test data of the subjects include the test data of the subjects answering questions containing various knowledge points. After obtaining the target knowledge points contained in the target questions, the historical test data are filtered based on the target knowledge points, and the questions in the test data containing the target knowledge points are used as candidate questions. Then, based on the correct answer rate of the candidate questions, the historical test accuracy rate for generating grade points is obtained. Specifically, the candidate questions may only contain the target knowledge points, or may contain the target knowledge points and other knowledge points. For example, "It is known that a, b, and c are three consecutive columns of an arithmetic progression, and a+b+c=21. What are the factors of b?", that is, it contains both the knowledge points of "arithmetic progression" and "factors".

[0028] In a specific implementation scenario, the subject's grade point ranges from 0 to 1, and the historical learning progress and historical question-solving accuracy are assigned corresponding weights, and the grade point is calculated. For example, the subject's historical learning progress is 0.5, the historical question-solving accuracy is 0.8, and the weight of the historical learning progress is set equal to the weight of the historical question-solving accuracy, both of which are 0.5, then the subject's grade point is calculated to be 0.65. It should be noted that the above embodiment is only a possible implementation method, and the calculation method of the grade point is not limited in this application.

[0029] In one implementation scenario, the assessment results also include the test result data and test process data of the subjects' answers to the target questions. For example, the test result data includes any one of all correct answers, partially correct answers, and all incorrect answers. It can be understood that when the standard answer to the target question includes several problem-solving steps, or there are multiple questions to be answered in the target question, the test result data is any one of all correct answers, partially correct answers, and all incorrect answers. In some implementation scenarios, the test result data also includes all the answers of the subjects to the target questions.

[0030] In a specific implementation scenario, the test-taking process data includes the time it takes the subject to answer the target question, the relevant analytical materials viewed during the problem-solving process, etc. For example, while the subject is answering the target question, the subject is provided with relevant knowledge point videos, knowledge point cards, interactive analysis of question-knowledge points, and other content. The subject can generate corresponding test-taking process data by viewing the relevant content.

[0031] In a specific implementation scenario, the test-taking process data can be used to measure the seriousness of the subjects in answering the target questions, such as judging whether there are random or quick answers, checking the answers in advance, and other behavioral data.

[0032] In a specific implementation scenario, the test-taking process data can be used to measure the degree of completion of the subject's learning process, such as determining whether the subject has watched a knowledge point video or knowledge card, or viewed interactive analysis and other behavioral data.

[0033] In one implementation scenario, target test point information contained in a target question is obtained, and the target test point information includes at least one of a knowledge graph about the target test point, a degree of difficulty of the target test point, and a distribution of test questions about the target test point, and a difficulty score of the target question is generated based on the target test point information.

[0034] In a specific implementation scenario, the knowledge graph of the target test point in the target test point information is used to describe the sequential relationship of different test point knowledge. For example, test point A is the predecessor of test point B, which means that mastering test point A helps to better master test point B. For another example, each knowledge point is divided into one of "basic knowledge point", "advanced knowledge point" and "advanced knowledge point" based on the knowledge graph of the knowledge point.

[0035] In a specific implementation scenario, the distribution of test questions about the target test point includes attribute information such as test frequency, real test score distribution, and real test difficulty distribution.

[0036] In a specific implementation scenario, the difficulty score of the target question and the grade score of the subject are set to the same score range. For example, the grade score of the subject ranges from 0 to 1, and the difficulty score of the target question also ranges from 0 to 1.

[0037] Step S21: Based on the evaluation results and difficulty scores, target feedback sentences and target learning sentences are selected.

[0038] In an implementation scenario, when the evaluation results also include the test subject's answering result data and test process data of the target question, based on the difficulty score of the target question and the test subject's test result data and grade score, candidate feedback sentences are selected as target feedback sentences. The candidate feedback sentences include several encouraging sentences representing different levels of feedback and several critical sentences representing different levels of feedback. For example, the candidate feedback sentences include a first-level encouraging sentence "Please keep up the good work", a second-level encouraging sentence "You are very good and answered this question correctly", and a third-level encouraging sentence "Great! You solved a difficult problem". The candidate feedback sentences also include a first-level critical sentence "Please carefully understand the content provided in the question stem" and a second-level critical sentence "Please don't answer the question perfunctorily". It can be understood that the above embodiment is only a possible implementation method. Candidate feedback sentences representing the same feedback level may be expressed in different ways. The specific sentence content of the candidate feedback sentences and the number of candidate feedback sentences are not limited in this application. In addition, based on the difficulty score of the target question and the test-taking process data and grade points of the subjects, candidate learning sentences are selected as target learning sentences. The candidate learning sentences include at least one of a test-taking evaluation sentence and a learning suggestion sentence. For example, the test-taking evaluation sentence includes "Step 5 references an incorrect calculation formula, resulting in an incorrect final calculation result", and the learning suggestion sentence includes "Please watch the teaching video a about the formula to consolidate the learning content", etc. It can be understood that the above embodiment is only one possible implementation method, and the specific content of the target learning sentence is not limited in this application.

[0039] In a specific implementation scenario, candidate feedback sentences are annotated with corresponding feedback degree label pairs. For example, "You did a good job" and "You are awesome" are both annotated with the feedback degree label pair "Level 2 - Encouragement", and "You were too careless when doing the test" is annotated with the feedback degree label pair "Level 2 - Criticism".

[0040] In a specific implementation scenario, the learning suggestion sentence in the candidate learning sentence is associated with the relevant knowledge points involved in the data of the problem-solving process. For example, if it is detected that the relevant knowledge point involved in the data of the problem-solving process is a certain "basic knowledge point", the consolidation test questions about this "basic knowledge point" can be recommended to the subject in the learning suggestion sentence, or the learning video about the "advanced knowledge point" of this "basic knowledge point" can be recommended to the subject. Alternatively, when the subject fails to answer the target question correctly, the learning video, learning card, etc. about this "basic knowledge point" can be recommended to the subject in the learning suggestion sentence.

[0041] In a specific implementation scenario, the target benchmark value is determined based on whether the subject correctly answers the target question as represented by the test result data. For example, when the subject answers all the target questions correctly, the target benchmark value is set to a maximum value of 1. When the subject answers part of the target questions correctly, the target benchmark value is determined according to the proportion of the subject's correct answers. For example, the target question includes question 1 and question 2, the subject answers question 1 correctly and answers question 2 incorrectly, and the target benchmark value is set to 0.5. For another example, the answer to the target question includes 5 complete steps, and the subject correctly answers 3 complete steps, and the target benchmark value is set to 0.6. When all the target questions are answered incorrectly, the target benchmark value is set to a minimum value of 0.1. It should be noted that the value range of the target benchmark value is not limited in this application. A first weight is determined based on the difficulty score, and a second weight is determined based on the grade score. The first weight is proportional to the difficulty score, and the second weight is inversely proportional to the grade score. The target benchmark value is weighted based on the first weight and the second weight, respectively, to obtain a weighted sum value. For example, the grade score of the subject ranges from 0 to 1, and the difficulty score of the target question also ranges from 0 to 1. The higher the grade score, the stronger the learning ability of the subject, and the higher the difficulty score, the more difficult the target question. A candidate feedback sentence with the target feedback degree corresponding to the sum is selected as the target feedback sentence.

[0042] In a specific implementation scenario, a mapping relationship between the difficulty score and the first weight is set. For example, when the difficulty score is greater than 0.8 and not greater than 1, the mapped first weight is 1.5, when the difficulty score is greater than 0.6 and not greater than 0.8, the mapped first weight is 1.2, when the difficulty score is greater than 0.4 and not greater than 0.6, the mapped first weight is 1, when the difficulty score is greater than 0.2 and not greater than 0.4, the mapped first weight is 0.8, and when the difficulty score is greater than 0 and not greater than 0.2, the mapped first weight is 0.6.

[0043] In a specific implementation scenario, a mapping relationship between the grade point and the second weight is set. For example, when the grade point is greater than 0.8 and not greater than 1, the mapped second weight is 0.6; when the grade point is greater than 0.6 and not greater than 0.8, the mapped second weight is 0.8; when the grade point is greater than 0.4 and not greater than 0.6, the mapped second weight is 1; when the grade point is greater than 0.2 and not greater than 0.4, the mapped second weight is 1.2; when the grade point is greater than 0 and not greater than 0.2, the mapped second weight is 1.5.

[0044] In a specific implementation scenario, a mapping relationship is set between different numerical values ​​and candidate feedback sentences of different feedback degrees. After the sum value is calculated, the corresponding candidate feedback sentence is selected as the target feedback sentence based on the mapping relationship. For example, taking the aforementioned embodiment as an example, the target baseline value is 0.5, the difficulty score of the target question is 0.7, that is, the first weight is 1.2, the level score of the subject is 0.7, that is, the second weight is 0.8, and the calculated sum value is 1. The feedback degree label pair corresponding to the candidate feedback sentence corresponding to the value 1 is "Level 1 - Encouragement". Accordingly, if the calculated sum value is less than the preset encouragement threshold, such as 0.3, the selected candidate feedback sentence is a criticism sentence. When the calculated sum value is not less than the preset encouragement threshold, the selected candidate feedback sentence is an encouragement sentence, and the feedback degree changes accordingly with the change of the sum value.

[0045] It should be noted that the above embodiment is only one possible implementation method, and the method for generating the first weight and the second weight and the method for selecting the target feedback sentence are not limited in this application.

[0046] Step S31: Generate learning status comments based on the target feedback sentence and the target learning sentence.

[0047] In one implementation scenario, the learning comments are presented in natural language. Natural language is based on the natural, informal language system used by humans and can be used for communication, expressing opinions, conveying information, etc. It should be noted that the language type of the learning comments is not limited in this application, such as Chinese, English, French, etc.

[0048] In a specific implementation scenario, the output form of the learning status comments is not limited in this application, such as text, voice, pictures, etc.

[0049] In a specific implementation scenario, the learning status comments are also associated with knowledge content related to the target topic, such as learning videos, extended exercises, etc.

[0050] In an implementation scenario, based on the feedback degree label pair contained in the target feedback sentence and the knowledge point content contained in the target learning sentence, the learning status comments are polished and generated. For example, the target feedback sentence is "Please keep up the good work", the feedback degree label pair contained is "Level 1 - Encouragement", the target learning sentence is "Please continue to learn relevant courses about arithmetic series", and the knowledge point content contained is "Arithmetic series". The learning status comments generated after polishing are "The relevant knowledge about arithmetic series is a bit weak, please keep up the good work. It is recommended to watch the explanation of arithmetic series in the series course video."

[0051] The above scheme selects target feedback sentences and target learning sentences based on the test results of the subject and the difficulty score of the target question, and the target feedback sentence includes any one of the encouraging sentence and the criticizing sentence, and the target learning sentence is used to indicate the learning content of the subject, and the learning situation comments are generated based on the target feedback sentence and the target learning sentence, and the evaluation results representing at least the learning level of the subject and the difficulty score representing the difficulty of the target question are used as reference information for generating the learning situation comments, and the learning situation analysis can be performed based on multi-angle data to avoid the bias caused by judging the learning level of the subject only from whether the target question is answered correctly, and the learning situation comments containing the target feedback sentence and the target learning sentence can enhance the subject's enthusiasm for participating in learning, timely feedback the subject's learning situation and give corresponding instructions, and enhance the subject's learning effect. Therefore, the accuracy of the learning situation analysis of the subject can be improved.

[0052] See also Figure 3 , Figure 3 1 is a schematic diagram of a framework of an embodiment of a learning situation comment generation model 30 in the learning situation analysis method of the present application. Figure 3 As shown, the learning status comment generation model 30 includes a difficulty score generation network 31, a level score generation network 32, a sentence generation network 33 and a large language model 34, and constructs a prompt text based on the target feedback sentence and the target learning sentence; wherein the prompt text is used to indicate the comment generation style of the large language model 34, and to adjust the knowledge text about the learning direction of the subject, and send the prompt text to the large language model 34, and the target sentence output by the large language model 34 serves as the learning status comment of the subject.

[0053] In an implementation scenario, the large language model 34 can understand human natural language input, and obtain keyword information about the input text and the logical relationship between each element according to the input text, and generate semantically related output. For specific principles, please refer to more technical details of the large language model, which will not be repeated here. For example, the target feedback sentence is "great", the target learning sentence is "the mastery rate of the test points and the correct answer rate are both high", and the constructed prompt text is "please combine [great], try to polish the [tone style] of the learning situation comment, and generate a learning situation comment about the current target learning sentence [the specific content of the current target learning sentence]", etc. The generated learning situation comment is "Dear classmate, you have a thorough grasp of the test points of this unit, and the correct answer rate is very high, which shows that your foundation is very solid and worthy of praise! It is recommended to continue to learn the advanced test point content about the current test point", so as to utilize the general understanding ability of the large language model 34 to generate learning situation comments. It should be noted that the large language model 34 can include but is not limited to: LLAMA, Bloom and other open source large models, and the network architecture of the large language model is not limited here.

[0054] The above scheme selects target feedback sentences and target learning sentences based on the test results of the subject and the difficulty score of the target question, and the target feedback sentence includes any one of the encouraging sentence and the criticizing sentence, and the target learning sentence is used to indicate the learning content of the subject, and the learning situation comments are generated based on the target feedback sentence and the target learning sentence, and the evaluation results representing at least the learning level of the subject and the difficulty score representing the difficulty of the target question are used as reference information for generating the learning situation comments, and the learning situation analysis can be performed based on multi-angle data to avoid the bias caused by judging the learning level of the subject only from whether the target question is answered correctly, and the learning situation comments containing the target feedback sentence and the target learning sentence can enhance the subject's enthusiasm for participating in learning, timely feedback the subject's learning situation and give corresponding instructions, and enhance the subject's learning effect. The learning situation comments generated by polishing based on the large language model can further enhance the subject's enthusiasm. Therefore, the accuracy of the learning situation analysis of the subject can be improved.

[0055] See also Figure 4 , Figure 4 This is a flow chart of another embodiment of the present application. Specifically, it may include the following steps: Step S12: Obtain the test results of the subject and the difficulty score of the target question.

[0056] In the disclosed embodiment, the learning situation analysis result includes at least the updated grade score of the subject and the updated difficulty score of the target question.

[0057] In one implementation scenario, the same target question may be selected by multiple subjects to answer. When the target question is not selected by any subject to answer, an initial difficulty score is set. Exemplarily, the initial difficulty score may be set based on a person skilled in the art, or may be determined based on the average difficulty score of all questions in the question bank, which will not be given one by one here.

[0058] In a specific implementation scenario, several questions are stored in a question bank, and the subjects select a question in the question bank as the target question, and the difficulty score of each question in the question bank is dynamically adjusted based on the selected answers of several subjects. It can be understood that when a question is selected as the target question by a sufficient number of subjects and answered, the difficulty score of the target question will stabilize within a certain difficulty score range, so that the difficulty score of each question can accurately represent the average difficulty, thereby improving the accuracy of the question difficulty score, and thus improving the accuracy of the learning situation analysis of the subjects.

[0059] In a specific implementation scenario, in response to the subject selecting any question in the question bank, the selected question is determined as the target question, and the subject's latest grade score and the difficulty score of the target question are obtained. It should be noted that the method for the subject to select the target question is not limited in this application.

[0060] Step S22: Based on the subject's grade score and the difficulty score of the target question, obtain the predicted score rate of the subject answering the target question, and obtain the actual score rate of the subject answering the target question.

[0061] In one implementation scenario, the rating and difficulty scores are obtained based on the Elo rating system, which is an evaluation method for measuring the level of various game activities. In this solution, the subject and the target question are regarded as a pair of game objects, and the subject's ability to answer questions is a normally distributed random variable. For example, the subject's ability to answer questions is affected by factors such as knowledge reserves and answering status and presents a normal distribution. The Elo rating system uses the average value of the random variable to represent the true level of the subject, that is, the rating is used to represent the learning ability of the subject. The difficulty of the target question is also a normally distributed random variable. The Elo rating system uses the average value of the random variable to represent the true difficulty of the target question, that is, the rating is used to represent the difficulty of the target question. The above solution regards the subject and the target question as a pair of game objects based on the Elo rating system, and dynamically adjusts the subject's rating and the difficulty of the target question to improve the accuracy of the subject's learning situation analysis.

[0062] In a specific implementation scenario, the grade points of the subjects and the difficulty points of the target questions use corresponding scoring standards. For example, the grade points range from 0 to 100, the difficulty points range from 0 to 100, and the grade point distribution trend of all subjects is similar to the difficulty point distribution trend of all target questions.

[0063] In one implementation scenario, based on the subject's grade score and the difficulty score of the target question, the predicted score rate of the subject's answer to the target question can be obtained. It can be understood that the Elo grade point system can make predictions based on the subject's grade score and the difficulty score of the target question to obtain the predicted score rate of the subject's answer to the target question. When the subject's grade score is greater than the difficulty score of the target question, the predicted score rate is higher, indicating that the subject is more likely to answer the target question correctly. When the subject's grade score is not greater than the difficulty score of the target question, the predicted score rate is lower, indicating that the subject is more likely to answer the question incorrectly, and the size of the predicted score rate is also related to the difference between the grade score and the difficulty score.

[0064] In a specific implementation scenario, the predicted score rate can be predicted based on the set score of the target question. For example, the score of a target question is 10 points. Based on the grade score of the subject and the difficulty score of the target question, it is predicted that the subject can get 6 points when answering the target question, and the predicted score rate when the subject answers the target question is determined to be 0.6. It should be noted that the above embodiment is only one possible implementation method, and for the sake of brevity, it will not be repeated here one by one.

[0065] In a specific implementation scenario, the Elo rating system is provided with a target parameter for characterizing the test subject's ability to answer questions, a first difference between the difficulty score and the rating score is obtained, and a target parameter determined for the test subject based on the distribution interval of the Elo rating system is obtained, and a predicted score rate is obtained based on the first difference and the target parameter. For details, please refer to the following formula:

[0066] In formula (1), Characterize the grade of the subject, Represents the difficulty score of the target question, Characterizing target parameters. It should be noted that the specific values ​​of the target parameters involved in the above formula are only one possible implementation method and are not limited in this application.

[0067] In one implementation scenario, the actual score rate of the subject answering the target question is obtained. For example, the actual score rate of the subject answering the target question correctly is 1, and the actual score rate of the subject answering the target question incorrectly is 0. Alternatively, the score of a target question is 10 points, and the subject gets 6 points when answering the target question, and the actual score rate of the subject when answering the target question is determined to be 0.6. It should be noted that the above embodiment is only one possible implementation method, and for the sake of brevity, it will not be repeated here one by one.

[0068] In a specific implementation scenario, after the subject confirms that the answer is completed, the actual score rate is obtained based on the subject's answer, and the predicted score rate is not affected by the result of this answer.

[0069] Step S32: Based at least on the predicted score rate and the actual score rate, update at least one of the level score and the difficulty score.

[0070] In one implementation scenario, after obtaining the predicted score rate and the actual score rate, and before updating the ranking points, a first adjustment coefficient for characterizing the score update step of the ranking points is obtained. Exemplarily, the first adjustment coefficient is 64, 32, etc., and examples are not given one by one here.

[0071] In a specific implementation scenario, after obtaining the first adjustment coefficient, the first adjustment score is determined based on the predicted score rate, the actual score rate and the first adjustment coefficient, and the grade score is updated based on the first adjustment score. In the above scheme, the subject's answer to each target question will be reflected in the change of the grade score, and the subject's grade score will be dynamically updated based on the different difficulty levels of each target question, so the accuracy of the subject's learning situation analysis can be improved.

[0072] In a specific implementation scenario, based on the total number of answers by the subject within a preset period, the first cumulative number of questions answered by the subject in the preset period is determined, and based on the first cumulative number of questions, the first adjustment coefficient is determined. The preset period can be found in the detailed description in the aforementioned embodiment, and for the sake of brevity, it will not be repeated here. The above scheme determines the first adjustment coefficient based on the first cumulative number of questions answered by the subject within the preset period, and the update step of the grade points can be dynamically adjusted to improve the accuracy of the learning situation analysis.

[0073] In a specific implementation scenario, based on whether the first cumulative number of questions reaches a preset question number threshold, a formula for determining the first adjustment coefficient based on the first cumulative number of questions is determined. For example, in order to allow the subject to quickly update the grade to the average level of learning ability in the early stage of a preset period, the preset question number threshold is set to 32 questions. When the first cumulative number of questions is not greater than 32 questions, the first adjustment coefficient is calculated using the following formula:

[0074] In formula (2), Characterizes the first adjustment coefficient, Represents the current question number. When the first cumulative number of questions is greater than 32, the first adjustment coefficient is determined to be 16. The above scheme, based on the number interval of the first cumulative number of questions, sets different first adjustment coefficients to obtain a determination method. When the cumulative number of questions answered by the subject is not greater than the preset question threshold, the subject's grade point can be quickly adjusted to near the average learning ability value in the early stage of the preset period. When the cumulative number of questions answered by the subject is greater than the preset question threshold, in order to avoid the subject from doing a large number of questions in order to improve the grade point, the first adjustment coefficient is set to an appropriate value to improve the accuracy of the learning situation analysis.

[0075] In a specific implementation scenario, a second difference between the actual score rate and the predicted score rate is obtained, and a first adjustment score is obtained based on the product of the second difference and the first adjustment coefficient. For details, please refer to the following formula:

[0076] In formula (3), The grade score representing the updated subject, Characterizes the grade score of the subject before the update, Characterizes the first adjustment coefficient, Represents the true score rate, Characterizes the prediction score rate, Represents the first adjustment score. In the above scheme, the subject's answer to each target question will be fed back in the change of grade points, and based on the different difficulty levels of each target question, the subject's grade points are dynamically updated, so the accuracy of the subject's learning situation analysis can be improved. When the difficulty score of the target question answered by the subject is greater than the subject's own grade point value, the reward for answering correctly is larger, that is, the increase in the grade point value is larger, and the penalty for answering incorrectly is smaller, that is, the decrease in the grade point value is smaller. On the contrary, when the difficulty score of the target question answered by the subject is less than the subject's own grade point value, the reward for answering correctly is smaller, that is, the increase in the grade point value is smaller, and the penalty for answering incorrectly is larger, that is, the decrease in the grade point value is larger. This helps to encourage the subject to continuously challenge questions that are slightly higher than his current learning ability, so as to continuously improve his own ability, so as to participate in learning more actively.

[0077] In a specific implementation scenario, the first adjustment coefficient is determined based on the total number of answers of the subject within a preset period, and the preset period includes several sub-periods. For example, the preset period is a semester, and each complete week from Monday to the weekend is a sub-period. The first adjustment score is also related to the number of questions answered by the subject in each sub-period. The sub-period to which the subject currently answers the target question is determined as the target sub-period. Based on the total number of answers in the target sub-period, the second cumulative number of questions when the subject currently answers the target question in the target sub-period is determined. Based on the grade and the second cumulative number of questions, the first preset coefficient is determined, and the first preset coefficient represents the score adjustment step of the first adjustment score. The first adjustment score is determined based on the product of the second difference, the first adjustment coefficient and the first preset coefficient. For example, one day is a sub-period. If the number of questions answered on that day is not greater than 8, the first preset coefficient is determined to be 1.15. If the number of questions answered on that day is greater than 8 and not less than 16, the first preset coefficient is determined to be 1. If the number of questions answered on that day is greater than 16, the first preset coefficient is determined to be 0.85. The above scheme helps to ensure that the subjects answer the target questions in each sub-period and avoid occasional cramming to develop good study habits.

[0078] In a specific implementation scenario, based on the mapping relationship between the subject's grade points and the grade set, the target grade of the subject is determined. The grade set includes several candidate grades. For details, please refer to the following Table 1:

[0079] Table 1 As shown in Table 1, based on the range of ability points, the target level and sub-level are used to represent the learning ability of the subjects, making the display of learning ability more concrete, so as to motivate the subjects to actively update the target level and improve the learning effect of the subjects.

[0080] See also Figure 5 , Figure 5 Schematic diagram of an embodiment of the target level distribution of subjects in the learning situation analysis method of the present application. Figure 5 As shown, the target grade distribution of the subjects presents a normal distribution trend, that is, the subjects' answers to each target question will be reflected in the changes in the grade points, and based on the different levels of difficulty of each target question, the subjects' grade points will be dynamically updated, thus improving the accuracy of the subjects' learning situation analysis.

[0081] In a specific implementation scenario, based on the current target level of the subject, a second preset coefficient is determined, the second preset coefficient represents the score adjustment step of the first adjustment score, and the first adjustment score is determined based on the product of the second difference, the first adjustment coefficient and the second preset coefficient.

[0082] In a specific implementation scenario, based on subjects with different target levels, the corresponding second preset coefficient is matched. For example, referring to Table 1, subjects with target levels of "D" and "E" have a corresponding second preset coefficient of "1.15", subjects with target levels of "B" and "C" have a corresponding second preset coefficient of "1.05", subjects with target levels of "A" have a corresponding second preset coefficient of "0.95", and subjects with target levels of "S" have a corresponding second preset coefficient of "0.85". The above scheme, according to the subject's grade adjustment strategy, reduces the difficulty of grade improvement and target grade improvement for subjects with weaker learning ability, and conversely increases the difficulty of grade improvement and target grade improvement for subjects with stronger learning ability.

[0083] In a specific implementation scenario, before updating at least one of the grade score and the difficulty score based on at least the predicted score rate and the actual score rate, a second adjustment coefficient is obtained, the second adjustment coefficient represents the score update step of the difficulty score, the second adjustment score is determined based on the predicted score rate, the actual score rate and the second adjustment coefficient, and the difficulty score is updated based on the second adjustment score. The above scheme performs a dynamic difficulty assessment based on the subject's answer to each target question to continuously update the difficulty score of each question, so that the difficulty score of each question can accurately represent the average difficulty, improve the accuracy of the question difficulty score, and thus improve the accuracy of the learning situation analysis of the subject.

[0084] In a specific implementation scenario, the difficulty score is updated based on the second adjustment score, and the specific formula may be as follows:

[0085] In formula (4), Represents the difficulty score of the updated target question, Represents the difficulty score of the target question before updating, Characterizes the second adjustment coefficient, Represents the true score rate, Characterizes the prediction score rate, The above scheme dynamically evaluates the difficulty of each target question answered by the subject, so as to continuously update the difficulty score of each question, so that the difficulty score of each question can accurately represent the average difficulty, improve the accuracy of the question difficulty score, and thus improve the accuracy of the learning situation analysis of the subject.

[0086] In a specific implementation scenario, the corresponding difficulty level is matched based on the difficulty score of the target question. For example, the questions are divided into different difficulty levels such as "simple", "basic" and "advanced" to avoid the deviation caused by manual labeling. This detailed difficulty division helps to meet the learning needs of subjects at different levels and improve learning efficiency.

[0087] In one implementation scenario, after updating at least one of the grade score and the difficulty score based on at least the predicted score rate and the actual score rate, the latest grade score of the subject and the latest difficulty score of each question are obtained, and based on the latest grade score, the difficulty score range when recommending questions to the subject is determined, and the recommended questions are determined based on the difficulty score range and the latest difficulty score of each question. The above scheme evaluates the learning ability of the subject based on the subject's latest grade score, and determines the difficulty of each question based on the latest difficulty score of each question, so as to improve the accuracy of the recommended questions and avoid the situation where the recommended questions are too difficult or too simple as much as possible.

[0088] In a specific implementation scenario, a recommended difficulty threshold is set. For example, the recommended difficulty threshold is 5, and the latest grade score of the subject is 65. Therefore, the difficulty score of the recommended question ranges from 60 to 70.

[0089] In a specific implementation scenario, recommended questions are displayed, and the order of displaying the recommended questions is determined based on the difference between the difficulty score of the recommended question and the grade score of the subject. For example, the recommended questions whose difficulty score is greater than the grade score are displayed at the front end, and are displayed on the recommended question display page in the order of the score difference from small to large. The above method, under the premise of motivating the subject to challenge more difficult target questions as much as possible, improves the quality of the recommended questions and actively promotes the subject to select and answer the recommended questions.

[0090] In a specific implementation scenario, after determining the recommended topic based on the difficulty score range and the latest difficulty score of each topic, in response to the subject's selection instruction for the recommended topic, the grade point update coefficient is obtained, and the recommended topic is selected as the new target topic, and the steps of obtaining the subject's grade point and the difficulty score of the target topic and subsequent steps are performed to update the subject's grade point and at least one of the difficulty score of the recommended topic, and the updated grade point is obtained based on the grade point update coefficient, the predicted score rate and the actual score rate. The above scheme encourages the subject to select the recommended topic for answering through the grade point update coefficient, which helps to help the subject perform personalized learning, and recommends better topics to the subject based on the subject's grade point and the difficulty score of each topic, so as to improve the accuracy of the subject's learning situation analysis. It can be understood that in order to encourage the subject to select the recommended topic for answering, when the subject selects the recommended topic for answering, the grade point update coefficient is not less than 1.

[0091] In another specific implementation scenario, after determining the recommended questions based on the difficulty score range and the latest difficulty score of each question, in response to the subject not triggering the selection instruction for the recommended question, the grade point update coefficient can be appropriately adjusted and the adjusted grade point update coefficient is not greater than 1. The above scheme prevents the subject from selecting target questions that are not suitable for his or her own learning ability to answer, thereby improving the efficiency of learning situation analysis.

[0092] In a specific implementation scenario, a historical answer database is constructed based on all the questions that the subject has answered. Before obtaining the subject's grade point and the difficulty score of the target question, it is detected whether the currently selected target question is the same as the question in the historical answer database. When there is a question that has been answered that is the same as the target question in the historical answer database, a pop-up window or other form is used to remind the subject that the question has been answered and whether it is necessary to continue answering. When the subject chooses to continue answering, the actual score rate of this answer is compared with the actual score rate of the previous answer. When the actual score rate of this answer is greater than the actual score rate of the previous answer, the subject's grade point is increased. When the actual score rate of this answer is equal to the actual score rate of the previous answer, the subject's grade point is kept unchanged. When the actual score rate of this answer is less than the actual score rate of the previous answer, the subject's grade point is reduced, and the increase step of the subject's grade point is not greater than the decrease step, so as to avoid the subject's behavior of brushing questions and motivate the subject to improve his memory ability for the questions that have been practiced.

[0093] In an implementation scenario, after updating at least one of the grade points and the difficulty points based on at least the predicted score rate and the actual score rate, the grade points obtained by the subjects after answering questions are updated respectively, and a change trend graph of the grade points of the subjects is generated, and the change trend graph is displayed. In the above method, the subjects can view the change trend graph used to characterize the change of their own learning ability in real time. When the change trend graph characterizes the gradual improvement of their own learning ability, the subjects can have a sense of accomplishment and satisfaction, and improve the incentive effect on the subjects. When the change trend graph characterizes the gradual reduction of their own learning ability, the learning attitude is actively adjusted to help the subjects learn more efficiently, based on the feedback of the grade points, the user experience of the subjects is enhanced, and the user's satisfaction and loyalty are improved. The change trend graph forms an intuitive grade point change curve graph by continuously tracking the subject's answering behavior, which not only helps the subjects to clearly see the trajectory of learning achievements and progress, stimulates learning enthusiasm, but also enables parents, teachers and other users to more accurately grasp the recent learning status of the subjects, and supervise and correct them in time.

[0094] In a specific implementation scenario, while the subject is answering the target question, the subject's grade score and the difficulty score of the target question are displayed in real time to motivate the subject to actively answer the target question and to increase the subject's enthusiasm for learning.

[0095] In a specific implementation scenario, the subject's ability score is compared with the school, region and other groups where the subject is located to show the relative position and performance of the subject in the group. This relative evaluation not only stimulates the subject's sense of competition, but also helps to reasonably establish the next stage of learning goals and promote active learning. For teachers and parents, it can provide a reference for mastering the actual level of the subject. It should be noted that the ability score of each subject needs to be collected under the authorization of the user, and based on the subject's confirmation and selection, the subject's relevant identity information in the school, region and other groups where the subject is located is disclosed or not, or the corresponding virtual identity information is set. Under the premise of disclosing the ability score, the identity information of the subject is not disclosed, which is not limited in this application.

[0096] In an implementation scenario, after each preset cycle, the subject's ability score is updated to maintain the initial grade score at the beginning of a new preset cycle. The subject can view the historical semester performance and is encouraged to continue learning and generate motivation for continuous improvement.

[0097] The above scheme uses grade points to represent the learning level of the subjects, and uses difficulty points to represent the difficulty of the questions. When the subjects select a certain question as the target question and answer it, the grade points of the subjects and the difficulty points of the target questions are obtained. Based on the grade points of the subjects and the difficulty points of the target questions, the predicted score rate of the subjects answering the target questions is obtained, and the actual score rate of the subjects answering the target questions is obtained. At least one of the grade points of the subjects and the difficulty points of the target questions is updated based on the predicted score rate and the actual score rate. The subjects' answers to each target question are fed back in the changes in the grade points, and based on the different difficulty levels of each target question, the grade points of the subjects are dynamically updated, thereby improving the accuracy of the learning situation analysis of the subjects.

[0098] See also Figure 6 , Figure 6 It is a schematic diagram of the framework of an embodiment of the learning situation analysis device 60 of the present application. The learning situation analysis device 60 includes an acquisition module 61 and a generation module 62. The acquisition module 61 is used to obtain the evaluation results of the subject and the difficulty score of the target question; wherein the evaluation results at least include a grade score representing the learning level of the subject; the generation module 62 is used to generate the learning situation analysis results of the subject based on the evaluation results and the difficulty score.

[0099] The above scheme generates the learning situation analysis results of the subjects based on the assessment results of the subjects and the difficulty scores of the target questions. The assessment results that at least represent the learning level of the subjects and the difficulty scores that represent the difficulty of the target questions are used as reference information for generating the learning situation analysis results. The learning situation analysis can be performed based on multi-angle data, thereby improving the accuracy of the learning situation analysis of the subjects.

[0100] In some disclosed embodiments, the learning situation analysis results include at least a learning situation comment on the subject, and the generation module 62 also includes a sentence selection module (not shown) for selecting a target feedback sentence and a target learning sentence based on the evaluation results and the difficulty score; wherein the target feedback sentence includes any one of an encouraging sentence and a criticizing sentence, and the target learning sentence is used to indicate the learning content of the subject; the generation module 62 also includes a comment generation module (not shown) for generating a learning situation comment based on the target feedback sentence and the target learning sentence.

[0101] In some disclosed embodiments, the evaluation results also include the test subject's answering result data and test process data of the target question, and the sentence selection module (not shown) also includes a candidate selection module (not shown) for selecting candidate feedback sentences as target feedback sentences based on the difficulty score of the target question and the test subject's test result data and grade score; wherein the candidate feedback sentences include a number of encouraging sentences representing different levels of feedback and a number of criticizing sentences representing different levels of feedback; and, based on the difficulty score of the target question and the test subject's test process data and grade score, candidate learning sentences are selected as target learning sentences; wherein the candidate learning sentences include at least one of a test evaluation sentence and a learning suggestion sentence.

[0102] In some disclosed embodiments, the candidate selection module (not shown) also includes a feedback sentence selection module (not shown), which is used to determine the target baseline value based on whether the subject represented by the test result data correctly answers the target question, and to determine the first weight based on the difficulty score and the second weight based on the grade score; wherein the first weight is proportional to the difficulty score and the second weight is inversely proportional to the grade score; the target baseline value is weighted based on the first weight and the second weight, respectively, to obtain a weighted sum; and a candidate feedback sentence of the target feedback degree corresponding to the sum is selected as the target feedback sentence.

[0103] In some disclosed embodiments, the acquisition module 61 also includes a grade point acquisition module (not shown), which is used to obtain the target knowledge points contained in the target questions, and to obtain the historical learning data and historical question-solving data of the subject; based on the historical learning data, the historical learning progress of the subject regarding the target knowledge points is determined, and based on the historical question-solving data, the historical question-solving accuracy of the subject regarding the target knowledge points is determined; based on the historical learning progress and the historical question-solving accuracy, the grade point of the subject is generated; wherein the historical learning progress is proportional to the grade point and the historical question-solving accuracy is proportional to the grade point.

[0104] In some disclosed embodiments, the acquisition module 61 also includes a difficulty score acquisition module (not shown), which is used to obtain target test point information contained in the target question; wherein the target test point information includes at least one of a knowledge graph about the target test point, the difficulty level of the target test point, and the distribution of test questions about the target test point; based on the target test point information, a difficulty score of the target question is generated.

[0105] In some disclosed embodiments, the generation module 62 also includes a large language generation module (not shown), which is used to construct a prompt text based on the target feedback sentence and the target learning sentence; wherein the prompt text is used to indicate the comment generation style of the large language model and to adjust the knowledge text about the learning direction of the subject; send the prompt text to the large language model; obtain the target sentence output by the large language model as the learning status comment of the subject.

[0106] In some disclosed embodiments, the learning situation analysis results include at least the subject's updated grade score and the updated difficulty score of the target question. The generation module 62 also includes a score rate acquisition module (not shown) for obtaining the predicted score rate of the subject's answer to the target question based on the subject's grade score and the difficulty score of the target question, and obtaining the actual score rate of the subject's answer to the target question; the generation module 62 also includes an update module (not shown) for updating at least one of the grade score and the difficulty score based on at least the predicted score rate and the actual score rate.

[0107] In some disclosed embodiments, the grade point and the difficulty point are obtained based on the Ello grade point system, and the score rate acquisition module (not shown) also includes a predicted score rate module (not shown), which is used to obtain a first difference between the difficulty point and the grade point, and obtain a target parameter determined for the subject based on the distribution range of the Ello grade point system; wherein the target parameter represents the subject's ability to answer questions; based on the first difference and the target parameter, the predicted score rate is obtained.

[0108] In some disclosed embodiments, the learning situation analysis device 60 also includes a first adjustment module (not shown), which is used to obtain a first adjustment coefficient before updating at least one of the grade point and the difficulty score based at least on the predicted score rate and the actual score rate; wherein the first adjustment coefficient represents the score update step of the grade point; updating at least one of the grade point and the difficulty score based at least on the predicted score rate and the actual score rate includes: determining a first adjustment score based on the predicted score rate, the actual score rate and the first adjustment coefficient; and updating the grade point based on the first adjustment score.

[0109] In some disclosed embodiments, the learning situation analysis device 60 also includes a second adjustment module (not shown), which is used to obtain a second adjustment coefficient before updating at least one of the grade score and the difficulty score based at least on the predicted score rate and the actual score rate; wherein the second adjustment coefficient represents the score update step of the difficulty score; updating at least one of the grade score and the difficulty score based at least on the predicted score rate and the actual score rate includes: determining the second adjustment score based on the predicted score rate, the actual score rate and the second adjustment coefficient; and updating the difficulty score based on the second adjustment score.

[0110] In some disclosed embodiments, the learning situation analysis device 60 also includes a question recommendation module (not shown), which is used to obtain the subject's latest grade score and the latest difficulty score of each question after updating at least one of the grade score and the difficulty score based on at least the predicted score rate and the actual score rate; based on the latest grade score, determine the difficulty score range when recommending questions to the subject; based on the difficulty score range and the latest difficulty score of each question, determine the recommended questions.

[0111] See also Figure 7 , Figure 7 : It is a schematic diagram of the framework of an embodiment of an electronic device 70 of the present application. The electronic device 70 includes a memory 71 and a processor 72. The memory 71 stores program instructions, and the processor 72 is used to execute the program instructions to implement the steps in any of the above-mentioned learning situation analysis method embodiments. For details, please refer to the aforementioned disclosed embodiments, which will not be repeated here. The electronic device 70 may specifically include but is not limited to: a server, a smart phone, a laptop computer, a tablet computer, a self-service machine, etc., which are not limited here.

[0112] Specifically, the processor 72 is used to control itself and the memory 71 to implement the steps in any of the above-mentioned learning situation analysis method embodiments. The processor 72 can also be called a CPU (Central Processing Unit). The processor 72 may be an integrated circuit chip with signal processing capabilities. The processor 72 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 72 can be implemented by an integrated circuit chip.

[0113] In the above scheme, the electronic device 70 generates a learning situation analysis result of the subject based on the assessment results of the subject and the difficulty score of the target question, and uses at least the assessment result representing the learning level of the subject and the difficulty score representing the difficulty of the target question as reference information for generating the learning situation analysis result. The learning situation analysis can be performed based on multi-angle data, thereby improving the accuracy of the learning situation analysis of the subject.

[0114] See also Figure 8 , Figure 8 1 is a schematic diagram of a framework of an embodiment of a computer-readable storage medium 80 of the present application. The computer-readable storage medium 80 stores program instructions 81 that can be executed by a processor, and the program instructions 81 are used to implement the steps in any of the above-mentioned learning situation analysis method embodiments.

[0115] In the above scheme, the computer-readable storage medium 80 generates a learning situation analysis result of the subject based on the assessment results of the subject and the difficulty score of the target question, and uses at least the assessment results that characterize the learning level of the subject and the difficulty score that characterizes the difficulty of the target question as reference information for generating the learning situation analysis result. The learning situation analysis can be performed based on multi-angle data, thereby improving the accuracy of the learning situation analysis of the subject.

[0116] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0117] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0118] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0119] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0120] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0121] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

[0122] If the technical solution of this application involves personal information, the product using the technical solution of this application has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution of this application involves sensitive personal information, the product using the technical solution of this application has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, the personal information processing rules are notified by obvious signs / information, and the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A learning situation analysis method, characterized in that: include: Obtaining the test results of the subject and the difficulty score of the target question; wherein the test results at least include a grade score representing the learning level of the subject; Based on the assessment results and the difficulty score, a learning situation analysis result of the subject is generated.

2. The method according to claim 1, characterized in that The learning situation analysis result at least includes a learning situation comment about the subject, and generating the learning situation analysis result of the subject based on the evaluation result and the difficulty score includes: Based on the evaluation result and the difficulty score, selecting a target feedback sentence and a target learning sentence; wherein the target feedback sentence includes any one of an encouraging sentence and a criticizing sentence, and the target learning sentence is used to indicate the learning content of the subject; The learning status comment is generated based on the target feedback sentence and the target learning sentence.

3. The method according to claim 2, characterized in that The evaluation result at least includes the test subject's answering result data and the test process data of the target question, and the selecting of the target feedback sentence and the target learning sentence based on the evaluation result and the difficulty score includes: Based on the difficulty score of the target question and the test result data and the grade score of the subject, a candidate feedback sentence is selected as the target feedback sentence; wherein the candidate feedback sentence includes a plurality of encouraging sentences representing different feedback degrees and a plurality of criticizing sentences representing different feedback degrees; And, based on the difficulty score of the target question and the test subject's test-taking process data and grade score, a candidate learning sentence is selected as the target learning sentence; wherein the candidate learning sentence includes at least one of a test-taking evaluation sentence and a learning suggestion sentence.

4. The method according to claim 3, characterized in that: The step of selecting a candidate feedback sentence as the target feedback sentence based on the difficulty score of the target question and the test result data and the grade score of the subject includes: Determine a target reference value based on whether the subject represented by the test result data correctly answers the target question, determine a first weight based on the difficulty score, and determine a second weight based on the grade score; wherein the first weight is proportional to the difficulty score, and the second weight is inversely proportional to the grade score; weighting the target reference value based on the first weight and the second weight respectively to obtain a weighted sum value; A candidate feedback sentence of the target feedback degree corresponding to the sum value is selected as the target feedback sentence.

5. The method according to claim 2, characterized in that: Obtaining the grade of the evaluation result includes: Obtaining target knowledge points included in the target question, and obtaining historical learning data and historical test-taking data of the subject; Based on the historical learning data, determine the historical learning progress of the subject with respect to the target knowledge point, and based on the historical question-solving data, determine the historical question-solving accuracy of the subject with respect to the target knowledge point; The grade score of the subject is generated based on the historical learning progress and the historical question-solving accuracy rate; wherein the historical learning progress is proportional to the grade score and the historical question-solving accuracy rate is proportional to the grade score.

6. The method according to claim 2, characterized in that Obtaining the difficulty score of the target question includes: Obtaining target test point information contained in the target question; wherein the target test point information includes at least one of a knowledge graph about the target test point, a degree of difficulty of the target test point, and a distribution of test questions about the target test point; Based on the target test point information, a difficulty score of the target question is generated.

7. The method according to claim 2, characterized in that The step of generating the learning status comment based on the target feedback sentence and the target learning sentence comprises: Based on the target feedback sentence and the target learning sentence, construct a prompt text; wherein the prompt text is used to indicate the comment generation style of the large language model and adjust the knowledge text about the learning direction of the subject; Sending the prompt text to the large language model; The target sentence output by the large language model is obtained as the learning status evaluation of the subject.

8. The method according to claim 1, characterized in that The step of generating the learning situation analysis result of the subject based on the evaluation result and the difficulty score includes: Based on the grade score of the subject and the difficulty score of the target question, obtaining the predicted score rate of the subject answering the target question, and obtaining the actual score rate of the subject answering the target question; At least one of the level score and the difficulty score is updated based on at least the predicted score rate and the actual score rate.

9. The method according to claim 8, characterized in that The grade point and the difficulty point are obtained based on the Elo grade point system, and the predicted score rate of the subject answering the target question based on the grade point of the subject and the difficulty point of the target question is obtained, including: Obtaining a first difference between the difficulty score and the rating score, and obtaining a target parameter determined for the subject based on the distribution interval of the Elo rating system; wherein the target parameter represents the subject's ability to answer questions; Based on the first difference and the target parameter, the prediction score rate is obtained.

10. The method according to claim 8, characterized in that Before updating at least one of the level score and the difficulty score based at least on the predicted score rate and the actual score rate, the method further includes: Obtaining a first adjustment coefficient; wherein the first adjustment coefficient represents the score update step of the level score; The updating of at least one of the level score and the difficulty score based at least on the predicted score rate and the actual score rate comprises: Determining a first adjustment score based on the predicted score rate, the actual score rate, and the first adjustment coefficient; Based on the first adjustment score, the rating score is updated.

11. The method according to claim 8, characterized in that Before updating at least one of the level score and the difficulty score based at least on the predicted score rate and the actual score rate, the method further includes: Acquire a second adjustment coefficient; wherein the second adjustment coefficient represents the score update step of the difficulty score; The updating of at least one of the level score and the difficulty score based at least on the predicted score rate and the actual score rate comprises: Determining a second adjustment score based on the predicted score rate, the actual score rate, and the second adjustment coefficient; Based on the second adjustment score, the difficulty score is updated.

12. The method according to any one of claims 8 to 11, characterized in that After updating at least one of the level score and the difficulty score based at least on the predicted score rate and the actual score rate, the method further includes: Obtaining the latest grade score of the subject and the latest difficulty score of each question; Determining a difficulty score range for recommending questions to the subject based on the latest grade score; Based on the difficulty score range and the latest difficulty score of each question, a recommended question is determined.

13. A learning situation analysis device, characterized in that: include: An acquisition module, used to acquire the test result of the subject and the difficulty score of the target question; wherein the evaluation result at least includes a grade score representing the learning level of the subject; A generation module is used to generate the learning situation analysis results of the subject based on the evaluation results and the difficulty score.

14. An electronic device, characterized in that: It comprises a memory and a processor coupled to each other, wherein the processor is used to execute program instructions stored in the memory to implement the learning situation analysis method described in any one of claims 1 to 12.

15. A computer-readable storage medium having program instructions stored thereon, characterized in that: When the program instructions are executed by the processor, the learning situation analysis method described in any one of claims 1 to 12 is implemented.

Images