Learning recommendation method and device based on wrong question analysis, equipment and storage medium

Abstract

The application relates to a learning recommendation method and device based on mistake analysis, equipment and a medium, and relates to information processing technology. The method comprises the following steps: obtaining mistake information; analyzing the mistake information based on a preset mistake model to obtain an analysis result; obtaining learning condition information and learning achievement information of a student; generating a learning strategy based on the analysis result, the learning condition information and the learning achievement information; obtaining learning time information and state information of the student; and recommending and displaying learning materials based on the learning time, the state information, the learning strategy and a preset material recommendation rule. The application has the effect of improving the learning effect and learning efficiency of students in consolidating knowledge points or deeply understanding knowledge points when learning mistakes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information processing, in particular to a learning recommendation method and device based on mistake analysis, equipment and a storage medium. BACKGROUND

[0002] With the continuous popularity of computers and the rapid development of information technology, classroom teaching and exercise are realized online, especially the online exercise process, which realizes the integration of doing, scoring and answering.

[0003] And the wrong question is often a manifestation of weak knowledge of students. Students organize the wrong questions into a corresponding mistake book during the learning process, and learn the content in the mistake book during review in order to better master knowledge points and problem solving skills during the learning process.

[0004] In related technologies, students record mistakes by recording mistakes in an online intelligent mistake book through image recognition, or record mistakes when students do online exercises, and classify mistakes. However, since the understanding and mastery of different students are different, students also learn systematically according to the mistake situation, so students may make the same mistake type again when reviewing mistakes, and the learning effect of knowledge consolidation or deep understanding of knowledge points is poor and the efficiency is low. SUMMARY

[0005] In order to improve the learning effect and learning efficiency of students in learning knowledge points based on mistakes, the present application provides a learning recommendation method and device based on mistake analysis, equipment and a storage medium.

[0006] In a first aspect, the present application provides a learning recommendation method based on mistake analysis, which adopts the following technical solution:

[0007] A learning recommendation method based on mistake analysis, comprising:

[0008] Obtaining mistake information; analyzing the mistake information based on a preset mistake model to obtain an analysis result; the analysis result includes mistake level, mistake type and mistake error reason;

[0009] Obtaining learning situation information and learning achievement information of a user; generating a first learning strategy based on the analysis result, the learning situation information and the learning achievement information;

[0010] Obtaining learning time information and state information of a user; recommending and displaying learning materials based on the learning time, the state information, the first learning strategy and a preset material recommendation rule.

[0011] By adopting the technical scheme, the wrong questions are classified by analyzing the error level, the wrong question type and the wrong question error cause, the learning progress of the user is intelligently recorded and analyzed, the learning strategy is generated for the knowledge points that are not firmly mastered according to the mastering degree and the learning situation of the user, the relevant knowledge points and historical wrong questions are recommended for review according to the learning time, the state information and the preset data recommendation rule, the learning effect and the learning efficiency of the students in learning the knowledge points by focusing on the wrong questions are improved, and the possibility of the user blindly reviewing the wrong questions and the related knowledge points is reduced.

[0012] Optionally, the wrong question information is analyzed based on a preset wrong question model to obtain an analysis result, and the analysis result includes:

[0013] The wrong question information is subjected to feature recognition to obtain feature information, and the feature information includes a keyword of the wrong question.

[0014] The wrong question original question is obtained based on the keyword of the wrong question, and the wrong question original question is preset with a test question level label and a test question type label.

[0015] The user input self-evaluation information is obtained, and the self-evaluation information includes at least one or more of concept ambiguity, thought error, test question error, operation error, carelessness and other errors.

[0016] The historical wrong question information of the same test question level and test question type as the wrong question information is obtained, and the historical wrong question information includes a wrong question quantity, teacher evaluation information, error time information and review degree information.

[0017] The wrong question error cause is analyzed based on the historical wrong question information, the self-evaluation information and a preset error cause analysis model.

[0018] By adopting the technical scheme, the original question corresponding to the wrong question is found and the test question level label and the test question type label are obtained by feature recognition of the wrong question information, so that the test question level and the test question type of the wrong question information are more accurate, and the wrong question cause is comprehensively analyzed by using the user self-evaluation and the historical wrong question information of the same test question level and test question type, so that the wrong question error cause is more comprehensive.

[0019] Optionally, after the wrong question information is obtained and the wrong question information is analyzed based on a preset wrong question model to obtain an analysis result, the method further includes:

[0020] Different wrong question storage models are created in advance according to different test question types.

[0021] store the wrong question information into corresponding storage models according to the analysis result; wherein each storage model is provided with a preset query algorithm based on the type of the test question, when the wrong question information needs to be queried, a query request is generated, the calculation engine routes the query request to the corresponding storage model, the storage model calls the corresponding preset query algorithm based on the query request, and the wrong question information to be queried is queried based on the query algorithm.

[0022] By adopting the above technical solution, the wrong question information is stored in the corresponding storage model according to the analysis result, the wrong questions are intelligently recorded, and the wrong questions are intelligently classified according to knowledge points, when the wrong questions are searched, the wrong question information is queried by using the query algorithm previously set in each storage model, and the efficiency of wrong question query and recommendation is improved.

[0023] Optionally, the first learning strategy is generated based on the analysis result, the learning condition information and the learning score information, and the first learning strategy includes:

[0024] The learning information associated with the analysis result is obtained based on the analysis result, and the learning information includes knowledge point information;

[0025] The learning condition information, the learning score information and the learning information associated with the analysis result are input into a preset correlation strength learning model to obtain learning weight information of each knowledge point;

[0026] The preset correlation strength information between the knowledge points associated with the analysis result is obtained;

[0027] The first learning strategy is generated based on the correlation strength information and the learning weight information of each knowledge point.

[0028] By adopting the above technical solution, the different degrees of user's mastery of related knowledge points are known according to the analysis result of the wrong questions, the learning method is provided for the user according to the correlation degree and the learning weight of each knowledge point, the weak knowledge points and the wrong questions are more accurately provided for the user at the present stage, the weak knowledge points are matched with the dynamic personalized adaptive learning scheme, and the user can efficiently learn.

[0029] Optionally, the learning material is recommended based on the learning time information, the state information, the first learning strategy and a preset material recommendation rule, and the learning material includes:

[0030] The first learning strategy includes a learning time proportion of learning knowledge points; and the state information includes at least one combination mode of a knowledge point review state and a wrong question practice state.

[0031] The learning time information is divided according to the learning time proportion to obtain the learning time of each knowledge point.

[0032] recommend learning materials based on the learning time of each knowledge point, the state information, and a preset learning material recommendation rule;

[0033] recommend learning materials based on the learning time of each knowledge point, the state information, and a preset learning material recommendation rule includes:

[0034] The historical wrong questions are pre-set with recommended values generated according to the number of review times, review time, and learning results of the user;

[0035] If the state information is a wrong question practice state, recommend wrong questions based on the learning time of each knowledge point and the recommended value of the historical wrong questions;

[0036] If the state information is a knowledge point review state, recommend the learning form of the knowledge point based on the learning time of each knowledge point; the learning form of the knowledge point includes knowledge point video explanation, knowledge point concept, knowledge point excellent learning notes, and knowledge point mind map.

[0037] By using the above technical solution, different learning methods are recommended for the user according to different state information and time of the user, so that the user can use fragmented time for personalized adaptive learning or long-term systematic targeted learning. Through different learning time, learning methods are recommended for the user and the learning time is accurately divided, and the learning efficiency of the user is improved.

[0038] Optionally, the wrong question information includes wrong question time, and the method further includes:

[0039] generate a second learning strategy based on the wrong question time, wrong question review degree, and a preset wrong question review rule, and send a prompt information;

[0040] generate learning content based on the second learning strategy and ensure that the learning content is stored in a wrong question review storage model; the learning content includes any one or a combination of several of teaching videos;

[0041] determine whether the learning content is processed within a preset time;

[0042] If not, determine whether the number of times that the user does not process the learning content exceeds a preset threshold;

[0043] If the number of times that the user does not process the learning content exceeds the preset threshold, send user review situation information prompts to a supervisor to supervise the user to learn the learning content.

[0044] By adopting the technical scheme, since the student will forget in the process of reviewing the wrong questions, the second learning strategy of the wrong question is formulated according to the wrong question review rule, the wrong question time and the wrong question review degree, the learning content specified according to the second learning strategy is generated and the user is automatically reminded to review the wrong question, if the user does not review the wrong question within the specified time, the user review situation is sent to the supervisor to supervise the user, and the exclusive personal wrong question review plan is established.

[0045] Optionally, the test question practice includes historical wrong questions and test questions of the same type as the historical wrong questions, and if the learning content is processed within the preset time, the method further includes:

[0046] After the user finishes the wrong questions, the test results are obtained;

[0047] If the test results of the historical wrong questions are correct, it is judged whether the review times of the historical wrong questions reach a preset review times threshold;

[0048] If the review times of the historical wrong questions reach the preset review times threshold, exercises of the same test question level and test question type as the historical wrong questions are recommended, and after the user answers correctly, the historical wrong questions are removed from the wrong question storage model;

[0049] If the review times of the historical wrong questions reach the preset review times threshold, the recommendation value of the historical wrong questions is reduced based on the review times of the historical wrong questions.

[0050] In a second aspect, the application provides a learning recommendation device based on wrong question analysis, which adopts the following technical scheme:

[0051] A learning recommendation device based on wrong question analysis includes:

[0052] A wrong question analysis module, a user obtains wrong question information; based on a preset wrong question model, the wrong question information is analyzed to obtain an analysis result; the analysis result includes test question level, test question type and wrong question error reason;

[0053] A learning strategy generation module is used to obtain student learning situation information and learning achievement information; based on the analysis result, the learning situation information and the learning achievement information, a learning strategy is generated;

[0054] A learning material recommendation module is used to obtain student learning time information and state information; based on the learning time, the state information, the learning strategy and a preset material recommendation rule, learning materials are recommended and displayed.

[0055] By adopting the technical scheme, the wrong questions are classified by analyzing the error level, the wrong question type and the wrong question error reason, the learning progress of the user is intelligently recorded and analyzed, the learning strategy is generated for the knowledge point which is not firmly mastered according to the mastering degree and the learning condition of the user, the relevant knowledge points and historical wrong questions are recommended for review according to the learning time, the state information and the preset data recommendation rule, the learning effect and the learning efficiency of the student in learning the knowledge points by focusing on the wrong questions are improved, and the possibility of the user blindly reviewing the wrong questions and the related knowledge points is reduced.

[0056] In a third aspect, the present application provides an electronic device, which adopts the technical scheme as follows:

[0057] An electronic device, comprising a memory and a processor, wherein the memory stores a computer program capable of being loaded and executed by the processor to perform the learning recommendation method based on wrong question analysis according to any one of the first aspect.

[0058] In a fourth aspect, the present application provides a computer readable storage medium, which adopts the technical scheme as follows:

[0059] A computer readable storage medium, which stores a computer program capable of being loaded and executed by the processor to perform the learning recommendation method based on wrong question analysis according to any one of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a flow diagram of the learning recommendation method based on wrong question analysis according to the embodiments of the present application.

[0061] Figure 2 is a structural block diagram of the learning recommendation device based on wrong question analysis according to the embodiments of the present application.

[0062] Figure 3 is a structural block diagram of an electronic device according to the embodiments of the present application. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme of the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0064] In addition, the term "and / or" in this document merely describes an association relationship of associated objects, which means that three relationships can exist, for example, A and / or B can represent three cases of existence of A alone, existence of A and B together, and existence of B alone. In addition, the character " / " in this document generally represents an "or" relationship between the front and rear associated objects unless otherwise specified.

[0065] An embodiment of the present application provides a learning recommendation method based on mistake analysis, which can be executed by an electronic device. The electronic device can be a server or a terminal device. The server can be a physical server, a server cluster composed of multiple physical servers, or a distributed system, and can also be a cloud server providing cloud computing services. The terminal device can be a tablet computer, a mobile phone, a desktop computer, or the like, but is not limited thereto. In the embodiment of the present application, the electronic device is a mistake intelligent terminal device.

[0066] The embodiment of the present application will be further described in detail below in combination with the accompanying drawings of the specification. As shown in the drawings, the main flow of the method is described as follows (steps S101-S103): Figure 1

[0067] In step S101, mistake information is acquired, and the mistake information is analyzed based on a preset mistake model to obtain an analysis result.

[0068] In the embodiment of the present application, the mistake intelligent terminal can start to collect image information of exercise books and test papers answered by the user in the starting mistake collection mode. The mistake collection mode can be a mode preset by the mistake intelligent terminal. After the user starts the mistake collection mode, the mistake intelligent terminal can input and collect the mistakes in the form of scanning, shooting, screen capture, handwriting, word batch import, and the like. In addition, the mistake collection mode is also connected with the exercise library system of the mistake intelligent terminal. The exercise library system is used for online exercises and learning of the user. The test questions in the exercise library system are stored in the cloud, and can be obtained through networking. The mistake intelligent terminal can also collect mistakes through the test questions answered incorrectly by the user in the exercise library system.

[0069] ​After the wrong question information is obtained, the wrong question information is analyzed. Specifically, first, the wrong question information is subjected to feature recognition to obtain feature information, which includes keywords of the wrong question. In the present embodiment, for the wrong question information entered by the user in the form of scanning, shooting, screen capture, etc., an optical character recognition technology can be used to scan and analyze the picture of the wrong question by using an optical character recognition program to automatically parse the keywords in the wrong question. For the wrong question made by the user in the exercise library system, the preset tags in the exercise library can be subjected to feature recognition to automatically parse the keywords in the wrong question. The conversion of the wrong question in the test paper or exercise into text or the automatic recording of the wrong question made by the user in the exercise library system improves the efficiency of the user in recording the wrong question. It should be noted that other character recognition technologies and image feature recognition technologies can also be used for analysis, and the present embodiment is not limited in this respect.

[0070] Then, the original question of the wrong question is obtained based on the keywords of the wrong question, and the original question of the wrong question is pre-provided with a test question level label and a test question type label. In the present embodiment, the keywords are matched with the exercise stored in the cloud to find the same exercise as the wrong question, and the keywords are used for matching to improve the search efficiency. After the user selects the same exercise as the wrong question, the test question level label and the test question type label in the exercise stored in the cloud are obtained. The test question level is obtained based on the big data analysis of the learning results of most users and is divided into “simple level test question”, “intermediate level test question”, “difficult level test question” and “super difficult level test question”. Each difficulty level is further divided into “easy-to-mistake question” and “typical question”. The test question type is divided according to each subject, and further divided according to each learning chapter of each subject and the test question type of each subject chapter, including but not limited to judgment question, single-choice question, multiple-choice question, fill-in-the-blank question, answer question and other types of questions.

[0071] Secondly, the user input self-evaluation information is obtained. In the present embodiment, after the wrong question intelligent terminal recognizes that the user has finished entering the wrong question, an evaluation page is displayed on the display interface of the wrong question intelligent terminal, and the evaluation page is used for the user to input the self-evaluation information, which includes at least one or more of concept ambiguity, thought error, question error, calculation error, carelessness and other errors, and the user can also mark the position where the wrong question is written wrong.

[0072] Then, the historical wrong question information of the same wrong question level and wrong question type as the wrong question information is obtained, and the historical wrong question information includes the number of wrong questions, teacher evaluation information, error time information and review degree information. In the present embodiment, the teacher evaluation can be the famous teacher evaluation corresponding to the wrong question obtained from the Internet, or can be the annotation of the teacher after assigning homework in the intelligent wrong question terminal. The review degree information is calculated based on the number of times of reviewing the wrong question by the user and the review result.

[0073] Finally, based on the history mistake information, the self-evaluation information and a preset error reason analysis model analysis, an error reason of the mistake is obtained. In the embodiment of the present application, the preset error reason analysis model is a neural network model, which can be a convolutional neural network model. The convolutional neural network model is trained by a large number of mistakes, self-evaluation, teacher evaluation, and related type mistake quantity information. Through feature recognition of the mistake information, the corresponding original question is found and the mistake grade label and the test type label are obtained, so that the test grade and test type of the mistake information are more accurate. The user self-evaluation and the historical mistake information of the same test grade and test type are used for comprehensive analysis of the mistake reason, so that the error reason of the mistake is more comprehensive.

[0074] After analyzing the mistake information based on the preset mistake model and obtaining the analysis result, the mistake information needs to be collected and recorded. Specifically, different mistake storage models are created in advance according to different test types.

[0075] The mistake information is stored in the corresponding storage model according to the analysis result. Each storage model is provided with a preset query algorithm based on the test type. When the mistake information needs to be queried, a query request is generated, and the computing engine routes the query request to the corresponding storage model. The storage model calls the corresponding preset query algorithm based on the query request, and queries the mistake information to be queried based on the query algorithm.

[0076] In the embodiment of the present application, for example, for the mistakes of mathematics subject, a storage model is established according to each subject chapter. The mistake intelligent terminal is also provided with corresponding buttons according to mistake types, mistake grades, mistake error times and other corresponding mistake labels. When the user needs to query the corresponding buttons of these types of mistakes, the corresponding storage model can be directly selected and queried.

[0077] By storing the mistake information in the corresponding storage model according to the analysis result, the mistakes are recorded intelligently, and the mistakes are intelligently classified according to knowledge points. When searching for mistakes, the query algorithm set in each storage model is used to query the mistake information, which improves the efficiency of mistake query and recommendation.

[0078] Step S102, obtaining learning situation information and learning achievement information of a student; generating a first learning strategy based on the analysis result, the learning situation information and the learning achievement information;

[0079] In the embodiments of the present application, the learning condition information is the learning mastery degree of the user learning each chapter, which is analyzed from the self-evaluation, the number of wrong questions, the wrong question level, the teacher evaluation information and the review time of the user, and specifically, the first weight can be allocated according to the self-evaluation of the mastery degree, the second weight can be allocated according to the number of wrong questions of each wrong question level, the third weight can be allocated according to the teacher evaluation information, the fourth weight can be allocated according to the review time recorded by the wrong question intelligent system, and the learning mastery degree of the user is calculated according to the weight of each condition and the preset mastery degree scoring standard, wherein the preset mastery degree scoring standard is obtained by big data analysis. The learning achievement information is the examination achievement of each subject each time and the grade ranking, daily learning examination achievement, etc., wherein the midterm examination, the final examination or the classroom practice can be uploaded by the user or the teacher through the achievement collection module.

[0080] Specifically, the generating the first learning strategy based on the analysis result, the learning condition information and the learning achievement information includes the following steps (steps S1021-S1024) (all shown in the figure):

[0081] In step S1021, learning information associated with the analysis result is obtained based on the analysis result, and the learning information includes knowledge point information.

[0082] In step S1022, the learning condition information, the learning achievement information and the learning information associated with the analysis result are input into a preset association strength learning model to obtain learning weight information of each knowledge point.

[0083] In step S1023, preset association strength information between knowledge points associated with the analysis result is obtained.

[0084] In step S1024, the first learning strategy is generated based on the association strength information and the learning weight information of each knowledge point.

[0085] In the embodiments of the present application, the basic knowledge points are complementary to each other, and they together constitute a big tree of knowledge. If the content of the later stage of the basic knowledge point is learned in a hurry without fully understanding the foregoing basic knowledge points, the user will continue to make mistakes in this type of wrong questions. Therefore, the associated knowledge points are obtained according to the analysis of the wrong question level, the wrong question type and other information, and the learning weight of each knowledge point and the association strength with the knowledge point are analyzed, so that the user can more comprehensively master the knowledge points corresponding to the wrong questions, and the weak knowledge points and the wrong questions are more accurately provided to the user at the present stage, the weak knowledge points are matched with the dynamic personalized adaptive learning scheme, and the user can learn efficiently.

[0086] In step S103, the learning time information and the state information of the student are obtained, and learning materials are recommended and displayed based on the learning time, the state information, the first learning strategy and a preset material recommendation rule.

[0087] In the embodiments of the present application, the first learning strategy includes a learning time proportion of learning knowledge points; the state information includes at least one combination of a knowledge point review state and a mistake question practice state; because the learning time of a student can be insufficient, for example, during a break or during fragmented time learning while waiting for class, there are many ways to learn knowledge points, and in the case of short learning time, the user can not know what learning method to choose for learning, so relevant learning materials are generated and recommended according to the learning time, the learning state and the first learning strategy, and the recommendation method of the learning materials includes, but is not limited to, any one combination of knowledge point video explanation, knowledge point concept, knowledge point excellent learning notes, knowledge point mind map and historical mistake questions.

[0088] Specifically, recommending and displaying learning materials based on the learning time, the state information, the first learning strategy and a preset material recommendation rule includes: dividing the learning time information according to the learning time proportion to obtain the learning time of each knowledge point; recommending learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule; in the embodiments of the present application, if the learning time is less than a preset learning threshold, a learning knowledge point is recommended according to the most important information of the learning weight information, and the learning form of the knowledge point is recommended according to the learning time and the state information.

[0089] Further, recommending learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule includes:

[0090] The historical mistake questions are pre-set with a recommended value generated according to the review times, review time and learning result of a user;

[0091] If the state information is a mistake question practice state, mistake questions are recommended based on the learning time of each knowledge point and the recommended value of the historical mistake questions; and the AJAX technology is used to present the questions and answers separately, so that the user can first independently contact the questions, think and answer, and after the answer is completed, the question and answer are displayed, the question and answer are separated, and the answer is avoided to interfere with the thinking of the questions. If the state information is a knowledge point review state, the learning form of the knowledge point is recommended based on the learning time of each knowledge point; the learning form of the knowledge point includes knowledge point video explanation, knowledge point concept, knowledge point excellent learning notes and knowledge point mind map.

[0092] By recommending different learning methods for the user according to different state information and time of the user, the user can utilize fragmented time for personalized adaptive learning or long-term systematic targeted learning, and the learning method is recommended for the user according to different learning time, and the learning time is accurately divided, so that the learning efficiency of the user is improved.

[0093] Since the student will forget in the process of reviewing the wrong questions, in order to improve the learning effect of the user, the user is automatically reminded to review the wrong questions.

[0094] Specifically, the learning recommendation method based on wrong question analysis further includes: first, generating a second learning strategy based on the wrong question time, the wrong question review degree and the preset wrong question review rule and issuing a prompt information; generating learning content based on the second learning strategy and ensuring into the wrong question review storage model; the learning content includes any one or several combinations of knowledge lecture, knowledge explanation video, excellent learning notes, mind map and test question practice; wherein, the preset wrong question review rule is formulated according to the forgetting curve theory and specific rules, and the student is automatically reminded to review the wrong questions at 24 hours, 48 hours, 144 hours and 720 hours after the user collects the wrong questions.

[0095] It is judged whether the learning content is processed within the preset time; if the learning content is not processed within the preset time, it is judged whether the number of times that the user does not process the learning content exceeds a preset threshold; if the number of times that the user does not process the learning content exceeds the preset threshold, the user review information is sent to the supervisor to supervise the user to learn the learning content.

[0096] If the learning content is processed within the preset time, the user finishes the wrong question, and the test result is obtained; if the test result of the historical wrong question is correct, it is judged whether the review times of the historical wrong question reach a preset review times threshold;

[0097] If the review times of the historical wrong question reach the preset review times threshold, the exercises of the same test question level and test question type as the historical wrong question are recommended, and after the user answers correctly, the historical wrong question is removed from the wrong question storage model; if the review times of the historical wrong question reach the preset review times threshold, the recommendation value of the historical wrong question is reduced based on the review times of the historical wrong question, and the historical wrong question mastered by the user is removed, so as to reduce the possibility of invalid test of the user.

[0098] By formulating the second learning strategy of the wrong question according to the wrong question review rule, the wrong question time and the wrong question review degree, generating the learning content specified according to the second learning strategy and automatically reminding the user to review the wrong question, if the user does not review the wrong question within the specified time, the user review information is sent to the supervisor to supervise the user, and the exclusive personal wrong question review plan is established.

[0099] The method classifies the wrong questions by analyzing the error level, error type and error reason of the wrong questions, intelligently records and analyzes the learning progress of the user, generates learning strategies for knowledge points that are not firmly mastered according to the mastering degree and learning situation of the user, and recommends relevant knowledge points and historical wrong questions for review according to the learning time, state information and preset material recommendation rules of the user, thereby improving the learning effect and learning efficiency of the students in consolidating knowledge points or deeply understanding knowledge points when learning wrong questions, and reducing the possibility of the user blindly reviewing wrong questions and related knowledge points.

[0100] The above embodiment introduces a learning recommendation method based on wrong question analysis from the perspective of method flow. The following embodiment introduces a learning recommendation device based on wrong question analysis from the perspective of virtual modules or virtual units. For details, see the following embodiment.

[0101] The embodiment of the present application provides a learning recommendation device based on wrong question analysis, as shown in the figure, the learning recommendation device based on wrong question analysis 200 comprises: Figure 2 The wrong question analysis module 201 obtains wrong question information of the user, analyzes the wrong question information based on a preset wrong question model, and obtains an analysis result;

[0102] The learning strategy generation module 202 is configured to obtain learning situation information and learning achievement information of the student, and generate a learning strategy based on the analysis result, the learning situation information and the learning achievement information;

[0103] The learning material recommendation module 203 is configured to obtain learning time information and state information of the student, and recommend and display learning materials based on the learning time, the state information, the learning strategy and a preset material recommendation rule.

[0104] As an optional implementation manner of the embodiment of the present application, the wrong question analysis module is specifically configured to:

[0105] perform feature recognition on the wrong question information to obtain feature information, wherein the feature information comprises keywords of the wrong question;

[0106] obtain a wrong question original question based on the keywords of the wrong question, wherein the wrong question original question is pre-provided with a test question level label and a test question type label;

[0107] obtain user input self-evaluation information, wherein the self-evaluation information comprises at least one or more of concept ambiguity, thought error, test question error, operation error, carelessness and other errors;

[0108]

[0109] ​Obtain historical mistake information of the same mistake level and mistake type as the mistake information, the historical mistake information including mistake quantity, teacher evaluation information, error time information and review degree information;

[0110] Analyze the historical mistake information, the self-evaluation information and a preset error cause analysis model to obtain a mistake error cause.

[0111] As an optional implementation of the embodiment of the present application, the learning recommendation device based on mistake analysis further includes a mistake storage module, specifically configured to:

[0112] After obtaining the mistake information and analyzing the mistake information based on the preset mistake model to obtain an analysis result, different mistake storage models are created in advance according to different test question types;

[0113] Store the mistake information into the corresponding storage model according to the analysis result; each storage model is provided with a preset query algorithm based on the test question type, when the mistake information needs to be queried, a query request is generated, the computing engine routes the query request to the corresponding storage model, the storage model calls the corresponding preset query algorithm based on the query request, and queries the mistake information to be queried based on the query algorithm.

[0114] As an optional implementation of the embodiment of the present application, the learning strategy generation module 202 is specifically configured to:

[0115] Obtain learning information associated with the analysis result based on the analysis result, the learning information including knowledge point information;

[0116] Input the learning situation information, the learning achievement information and the learning information associated with the analysis result into a preset correlation strength learning model to obtain learning weight information of each knowledge point;

[0117] Obtain preset correlation strength information between the knowledge points associated with the analysis result;

[0118] Produce a first learning strategy based on the correlation strength information and the learning weight information of each knowledge point.

[0119] As an optional implementation of the embodiment of the present application, the learning material recommendation module 203 is specifically configured to:

[0120] The first learning strategy includes a learning time proportion of learning knowledge points; the state information includes at least one combination mode of knowledge point review state and mistake practice state;

[0121] Divide the learning time information according to the learning time proportion to obtain the learning time of each knowledge point;

[0122] recommend learning materials based on the learning time of each knowledge point, the state information, and a preset learning material recommendation rule;

[0123] recommend learning materials based on the learning time of each knowledge point, the state information, and a preset learning material recommendation rule includes:

[0124] The historical wrong questions are pre-set with recommended values generated according to the number of review times, review time, and learning result of the user;

[0125] If the state information is a wrong question practice state, the wrong questions are recommended based on the learning time of each knowledge point and the recommended value of the historical wrong questions;

[0126] If the state information is a knowledge point review state, the learning form of the knowledge point is recommended based on the learning time of each knowledge point; the learning form of the knowledge point includes knowledge point video explanation, knowledge point concept, knowledge point excellent learning notes, and knowledge point mind map.

[0127] As an optional implementation of the embodiment of the application, the wrong question information includes wrong question time, and the learning recommendation based on wrong question analysis further includes a review prompt module, which is configured to:

[0128] generate a second learning strategy and issue a prompt information based on the wrong question time, wrong question review degree, and a preset wrong question review rule;

[0129] generate learning content based on the second learning strategy and ensure that the learning content is stored in a wrong question review storage model; the learning content includes any one or a combination of several of knowledge lectures, knowledge explanation videos, excellent learning notes, mind maps, and test question exercises;

[0130] determine whether the learning content is processed within a preset time;

[0131] If the learning content is not processed within the preset time, it is determined whether the number of times that the user does not process the learning content exceeds a preset threshold;

[0132] If the number of times that the user does not process the learning content exceeds the preset threshold, the user review situation information prompt is sent to a supervisor to supervise the user to learn the learning content.

[0133] As an optional implementation of the embodiment of the application, the test question exercise includes historical wrong questions and test questions of the same type as the historical wrong questions, and the learning recommendation device based on wrong question analysis further includes a wrong question removal module, which is configured to:

[0134] If the learning content is processed within the preset time, the test question result is obtained after the user finishes the wrong questions.

[0135] If the result of the historical error question is correct, it is determined whether the number of review times of the historical error question reaches a preset review time threshold value;

[0136] If the number of review times of the historical error question reaches the preset review time threshold value, an exercise of the same test question level and test question type as the historical error question is recommended, and after the user answers correctly, the historical error question is removed from the error question storage model;

[0137] If the number of review times of the historical error question reaches the preset review time threshold value, the recommendation value of the historical error question is reduced based on the number of review times of the historical error question.

[0138] In one example, the modules in any of the above apparatuses can be one or more integrated circuits configured to implement one or more of the above methods, such as one or more application specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0139] For another example, when the modules in the apparatuses can be implemented in the form of a processing element scheduler, the processing element can be a general purpose processor, such as a central processing unit (CPU) or other processor that can invoke programs. For another example, these modules can be integrated together to implement in the form of a system-on-a-chip (SOC).

[0140] In the present application, various messages / information / devices / network elements / systems / apparatuses / actions / operations / processes / concepts, etc. of various objects that can occur in the present application are named. It can be understood that these specific names do not constitute a limitation on the related objects, and the assigned names can be changed with factors such as scene, context or usage habits. The technical meaning of the technical terms in the present application should be mainly determined from the function and technical effect embodied / implemented in the technical scheme.

[0141] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, apparatus and module can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0142] Those skilled in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0143] Figure 3 An electronic device 300 according to an embodiment of the present application is shown in a structural block diagram.

[0144] As shown in Figure 3 , the electronic device 300 includes a processor 301 and a memory 302, and can further include one or more of an information input / output (I / O) interface 303 and a communication component 304.

[0145] The processor 301 is configured to control overall operations of the electronic device 300 to complete all or part of the steps of the learning recommendation method based on error analysis described above; the memory 302 is configured to store various types of data to support the operations of the electronic device 300, which can include, for example, instructions for operating any application or method on the electronic device 300, and application-related data. The memory 302 can be realized by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk.

[0146] The I / O interface 303 provides an interface between the processor 301 and other interface modules, which can be a keyboard, a mouse, a button, etc. The buttons can be virtual buttons or physical buttons. The communication component 304 is configured to test wired or wireless communication between the electronic device 300 and other devices. The wireless communication, for example, Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or one or more of them or a combination of them, and accordingly the communication component 304 can include a Wi-Fi component, a Bluetooth component, an NFC component.

[0147] The communication bus 305 can include a path for transmitting information between the components. The communication bus 305 can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 can be divided into an address bus, a data bus, a control bus, etc.

[0148] The electronic device 300 can be implemented by one or more ASICs (Application Specific Integrated Circuit), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), controllers, microcontrollers, microprocessors, or other electronic elements for performing the error analysis-based learning recommendation method according to the embodiments.

[0149] The electronic device 300 can include, but is not limited to, a mobile terminal such as a digital broadcast receiver, a PDA (Personal Digital Assistant), a PMP (Portable Multimedia Player), etc., and a fixed terminal such as a digital TV, a desktop computer, etc., and can also be a server, etc.

[0150] The computer-readable storage medium according to the embodiments of the present application will be described below. The computer-readable storage medium described below can be referred to in conjunction with the error analysis-based learning recommendation method described above.

[0151] The application further provides a computer readable storage medium, and the computer readable storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the learning recommendation method based on the error question analysis.

[0152] The computer readable storage medium can include a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media capable of storing program codes.

[0153] The term "comprises" or "comprising" or any other variant thereof is intended to cover a non-exclusive inclusion, so that a process, method, article or apparatus that comprises a list of elements does not only include those elements, but also other elements not expressly listed or inherent to such process, method, article or apparatus.

[0154] The above description is merely preferred embodiments of the present application and a description of the principles of the technology used. Those skilled in the art should understand that the application scope involved in the present application is not limited to the technical solutions formed by the specific combinations of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or equivalent features without departing from the above application concept. For example, the above features are replaced with the technical features applied in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. A learning recommendation method based on mistake analysis, characterized by, The method comprises the following steps: obtaining wrong question information; analyzing the wrong question information based on a preset wrong question model to obtain an analysis result; the analysis result comprises a test question level, a test question type and a wrong question error cause; obtaining learning condition information and learning achievement information of a user; generating a first learning strategy based on the analysis result, the learning condition information and the learning achievement information; obtaining learning time information and state information of the user; and recommending learning materials based on the learning time information, the state information, the first learning strategy and a preset material recommendation rule; the generation of the first learning strategy based on the analysis result, the learning condition information and the learning achievement information comprises: obtaining learning information associated with the analysis result based on the analysis result, wherein the learning information comprises knowledge point information; inputting the learning condition information, the learning achievement information and the learning information associated with the analysis result into a preset association strength learning model to obtain learning weight information of each knowledge point; obtaining preset association strength information between knowledge points associated with the analysis result; generating the first learning strategy based on the association strength information and the learning weight information of each knowledge point; the recommendation of learning materials based on the learning time information, the state information, the first learning strategy and a preset material recommendation rule comprises: the first learning strategy comprises a learning time proportion of a learning knowledge point; and the state information comprises at least one combination mode of a knowledge point review state and a wrong question practice state; dividing the learning time information according to the learning time proportion to obtain learning time of each knowledge point; recommending learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule; the recommendation of learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule comprises: a historical wrong question is provided with a recommended value generated according to a user review times, a review time and a learning result; if the state information is a wrong question practice state, a wrong question is recommended based on the learning time of each knowledge point and the recommended value of the historical wrong question; if the state information is a knowledge point review state, a learning form of the knowledge point is recommended based on the learning time of each knowledge point; the learning form of the knowledge point comprises a knowledge point video explanation, a knowledge point concept, a knowledge point excellent learning note and a knowledge point mind map.

2. The method of claim 1, wherein, the analysis of the wrong question information based on a preset wrong question model to obtain an analysis result comprises: performing feature recognition on the wrong question information to obtain feature information, wherein the feature information comprises a keyword of the wrong question; obtaining a wrong question original question based on the keyword of the wrong question, wherein the wrong question original question is provided with a test question level label and a test question type label; obtaining self-evaluation information input by a user, wherein the self-evaluation information comprises at least one or more of a concept ambiguity, a thought error, a test question error, an operation error, a careless mistake and other errors; obtaining historical wrong question information of the same wrong question level and wrong question type as the wrong question information, wherein the historical wrong question information comprises a wrong question quantity, teacher evaluation information, error time information and review degree information; Based on the historical mistake information, the self-evaluation information and a preset mistake reason analysis model analysis, a mistake error reason is obtained.

3. The method according to claim 1 or 2, characterized in that, In the acquisition mistake information; After analyzing the mistake information based on a preset mistake model to obtain an analysis result, the method further comprises: Different mistake storage models are created in advance according to different test question types; According to the analysis result, the mistake information is stored in the corresponding storage model; wherein each storage model is provided with a preset query algorithm based on the test question type setting, when the mistake information needs to be queried, a query request is generated, the calculation engine routes the query request to the corresponding storage model, the storage model calls the corresponding preset query algorithm based on the query request, and queries the mistake information to be queried based on the query algorithm.

4. The method of claim 1, wherein, The mistake information includes mistake time, and the method further comprises: Based on the mistake time, the mistake review degree and a preset mistake review rule, a second learning strategy is generated and a prompt information is sent out; Based on the second learning strategy, learning content is generated and ensured to be stored in the mistake review storage model; the learning content includes any one or a combination of several of knowledge lectures, knowledge explanation videos, excellent learning notes, mind maps and test question exercises; It is judged whether the learning content is processed within a preset time; If the learning content is not processed within the preset time, it is judged whether the number of times that the user does not process the learning content exceeds a preset threshold; If the number of times that the user does not process the learning content exceeds the preset threshold, user review information is sent to a supervisor to supervise the user to learn the learning content.

5. The method of claim 4, wherein, The test question exercise includes historical mistakes and test questions of the same type as the historical mistakes, and if the learning content is processed within a preset time, the method further comprises: After the user finishes the mistakes, a test result is obtained; If the test result of the historical mistake is correct, it is judged whether the review times of the historical mistake reach a preset review times threshold; If the review times of the historical mistake reach the preset review times threshold, exercises of the same test question level and test question type as the historical mistake are recommended, and after the user answers correctly, the historical mistake is removed from the mistake storage model; If the review times of the historical mistake do not reach the preset review times threshold, the recommendation value of the historical mistake is reduced based on the review times of the historical mistake.

6. A learning recommendation device based on mistake analysis, characterized by, It includes, A mistake analysis module, the user acquires mistake information; Based on a preset mistake model, the mistake information is analyzed to obtain an analysis result; the analysis result includes mistake level, mistake type and mistake error reason; A learning strategy generation module is used to acquire student learning information and learning achievement information; based on the analysis result, the learning information and the learning achievement information, a learning strategy is generated; A learning material recommendation module is used to acquire student learning time information and state information; based on the learning time, the state information, the learning strategy and a preset material recommendation rule, learning materials are recommended and displayed; The learning strategy generation module is specifically used for: Based on the analysis result, learning information associated with the analysis result is acquired, and the learning information includes knowledge point information; inputting the learning condition information, the learning achievement information and the learning information associated with the analysis result into a preset association strength learning model to obtain learning weight information of each knowledge point; obtaining preset association strength information between knowledge points associated with the analysis result; generating a first learning strategy based on the association strength information and the learning weight information of each knowledge point; The learning material recommendation module is specifically configured to: The first learning strategy includes a learning time proportion of learning knowledge points; and the state information includes at least one combination mode of a knowledge point review state and a mistake question practice state. dividing the learning time information according to the learning time proportion to obtain learning time of each knowledge point; recommending learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule; The recommendation of learning materials based on the learning time of each knowledge point, the state information and a preset material recommendation rule includes: The historical mistake questions are pre-set with a recommendation value generated according to a user review number, a review time and a learning result; if the state information is the mistake question practice state, recommending mistake questions based on the learning time of each knowledge point and the recommendation value of the historical mistake questions; if the state information is the knowledge point review state, recommending a learning form of the knowledge point based on the learning time of each knowledge point; the learning form of the knowledge point includes knowledge point video explanation, knowledge point concept, knowledge point excellent learning notes and knowledge point mind map.

7. An electronic device, comprising: comprising a processor coupled with a memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method of any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, comprising a computer program or instructions, when the computer program or instructions run on a computer, make the computer execute the method of any one of claims 1-5.

Images