An intelligent content and question recommendation system and method for adult self-study examinations

Abstract

The present invention relates to an intelligent adult self-study examination content and question recommendation system and method, belonging to the technical field of data processing. The method includes: based on the past adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions, establishing a first adult self-study examination knowledge point map corresponding to each region for different adult self-study examination majors; classifying based on the basic information, learning ability characteristic information and self-study records of multiple test users, determining multiple test users corresponding to each first adult self-study examination knowledge point map, and generating a corresponding second adult self-study examination knowledge point map; based on the basic information, learning ability characteristic information and classification results of the user to be recommended, determining multiple reference test users and the second adult self-study examination knowledge point map, and combining the self-study records of the user to be recommended to generate self-study examination recommended content and recommended questions, having the advantage of improving the accuracy of adult self-study examination content and question recommendation.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an intelligent adult self-study examination content and question recommendation system and method. Background Art

[0002] Self-taught higher education examinations, abbreviated as self-study examinations or self-taught exams, are national higher education exams mainly focusing on academic exams for self-study exam participants. It is a form of higher education combining individual self-study, social tutoring, and national exams. Self-study exam participants register and take exams at the time and place specified by provincial education examination institutions. After obtaining passing grades in one or more (including) courses, the provincial education examination institutions establish a student status management file for them. After self-study exam participants obtain all the passing grades required by the professional plan, complete other teaching practice tasks such as the graduation assessment or thesis defense of the host school, and pass the ideological and moral appraisal, they can obtain graduation certificates, which are recognized by the state. Self-study exam undergraduate graduates who meet the degree awarding conditions will be awarded a bachelor's degree by the host school with the right to confer degrees in accordance with relevant regulations. The main advantages of adult self-study exams include loose registration conditions, flexible study time, and high recognition of academic qualifications. The disadvantages are that the exam difficulty is relatively high, and candidates need to have strong self-study ability and time management ability. In addition, the tuition fees for self-study exam undergraduates are relatively low, but candidates need to arrange their own study and exams.

[0003] Recommending exam content and questions to students can effectively improve their learning effect when the number of questions to be done is determined. In the traditional way, this part is generally completed by the instructor. Based on their understanding of the students and the questions in their memory, the instructor selects appropriate exam content and questions for the students to practice. First, considering the number of questions the students do, it is difficult for the instructor to directly recommend a new question in real time every time a question is completed. Second, considering the limited memory of the instructor, it is difficult to remember the abilities of each student in various knowledge aspects and all the exam content and questions. Therefore, this recommendation method can only ensure the matching of the most important needs of the students and the exam content and questions that the instructor most wants to recommend, and it is difficult to ensure that the matching still holds after long-term use. Common recommendation algorithms are generally based on the two basic assumptions of "things that users like are similar" and "similar users like similar content", so as to recommend approximate content that users like and the content that similar users like. However, this scheme is not feasible for question recommendation. In the process of adult self-study, since most valuable exam content and questions are relatively difficult, students often do not like those exam content and questions that are truly helpful for learning. Therefore, the assumption based on liking cannot be applied to the recommendation of exam content and questions, and the reliability of such recommendation algorithms is naturally out of the question. General automated exam content and question recommendations are based on the classification and / or difficulty of exam content and questions. This method can ensure that the recommended exam content and questions meet the requirements of students in the general direction, but in fact, the recommendation accuracy is still very poor. First, the classification-based scheme is not completely reasonable. Questions often involve multiple aspects of knowledge, and forcing them into a certain category itself loses the recommendation accuracy. At the same time, since the difficulty is often marked for the questions, the matching for students' abilities can only be achieved at the question level, and the mutual influence of the difficulty between various parts within the exam content and questions is often ignored. Due to these two reasons, the data accuracy of its recommendation algorithm is poor. At the same time, traditional recommendation algorithms mostly determine exam content and questions completely based on the size of indicators, which leads to the situation that students are more likely to see the same exam content and questions at the same level. Repeating the study of the same exam content or questions cannot effectively improve the real level of students. Therefore, completely deterministic algorithms cannot meet the needs of students' self-study.

[0004] Therefore, there is a need to provide an intelligent adult self-study exam content and question recommendation system and method to improve the accuracy of adult self-study exam content and question recommendations. Summary of the Invention

[0005] The present invention provides an intelligent method for recommending content and questions for adult self-study examinations, including: obtaining previous adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions; based on the previous adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions, establishing a first adult self-study knowledge point map corresponding to each region for different adult self-study examination majors; obtaining basic information, learning ability characteristic information, and self-study records of multiple test users; classifying the multiple test users based on the basic information, learning ability characteristic information, and self-study records of the multiple test users to generate a classification result; determining multiple test users corresponding to each of the first adult self-study knowledge point maps based on the classification result; for each of the first adult self-study knowledge point maps, generating a second adult self-study knowledge point map corresponding to the first adult self-study knowledge point map based on the multiple test users corresponding to the first adult self-study knowledge point map; obtaining basic information and learning ability characteristic information of a user to be recommended; determining multiple reference test users and a second adult self-study knowledge point map corresponding to the user to be recommended from the multiple test users based on the basic information and learning ability characteristic information of the user to be recommended and the classification result; obtaining the self-study record of the user to be recommended; generating self-study examination recommended content and recommended questions corresponding to the user to be recommended based on the self-study record of the user to be recommended, the corresponding multiple reference test users, and the second adult self-study knowledge point map.

[0006] Further, establishing a first adult self-study knowledge point map corresponding to each region for different adult self-study examination majors based on the previous adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions includes: for each region, determining the examination knowledge points corresponding to different adult self-study examination majors in the region and the examination characteristics corresponding to each examination knowledge point based on the previous adult self-study examination papers corresponding to different adult self-study examination majors in the region; for each adult self-study examination major, clustering the multiple regions based on the examination knowledge points corresponding to the adult self-study examination major in each region and the examination characteristics corresponding to each examination knowledge point to determine multiple region clusters corresponding to the adult self-study examination major, and establishing a first adult self-study knowledge point map corresponding to the region cluster for the adult self-study examination major based on the examination knowledge points corresponding to the adult self-study examination major in the multiple regions included in each region cluster and the examination characteristics corresponding to each examination knowledge point.

[0007] Further, based on the previous years' adult self-study examination papers corresponding to different majors of adult self-study examinations in the region, determine the examination knowledge points corresponding to different majors of adult self-study examinations in the region and the examination characteristics corresponding to each examination knowledge point, including: for each major of adult self-study examination, construct a knowledge point library corresponding to the major of adult self-study examination; for each adult self-study examination paper, preprocess the adult self-study examination paper to generate a preprocessed adult self-study examination paper, and according to the knowledge point library corresponding to the major of adult self-study examination related to the adult self-study examination paper, extract knowledge points and knowledge point characteristics from the preprocessed self-study examination paper, where the knowledge point characteristics at least include the question proportion, question difficulty distribution, and question type distribution of each knowledge point; for each major of adult self-study examination, based on the knowledge points and knowledge point characteristics of the previous years' adult self-study examination papers corresponding to the major of adult self-study examination in the region, determine the examination knowledge points corresponding to the major of adult self-study examination in the region and the examination characteristics corresponding to each examination knowledge point.

[0008] Further, based on the examination knowledge points corresponding to each major of adult self-study examination in each region and the examination characteristics corresponding to each examination knowledge point, cluster the multiple regions to determine multiple region clusters corresponding to the major of adult self-study examination; based on the examination knowledge points corresponding to each major of adult self-study examination in each region, calculate the coincidence degree of the examination knowledge points corresponding to any two regions for the major of adult self-study examination; based on the examination characteristics corresponding to each examination knowledge point corresponding to each major of adult self-study examination in each region, calculate the similarity degree of the examination characteristics corresponding to any two regions for the major of adult self-study examination; based on the coincidence degree of the examination knowledge points corresponding to any two regions for the major of adult self-study examination and the similarity degree of the examination characteristics corresponding to any two regions for the major of adult self-study examination, calculate the region similarity degree corresponding to any two regions for the major of adult self-study examination; based on the region similarity degree corresponding to any two regions for the major of adult self-study examination, cluster the multiple regions to determine multiple region clusters corresponding to the major of adult self-study examination.

[0009] Further, obtain the learning ability characteristic information of the test users, including: determine multiple learning ability factors; obtain the ability test data of multiple test users; for each major of adult self-study examination, based on the ability test data of the multiple test users, determine the correlation coefficient between each learning ability factor and the major of adult self-study examination, and based on the correlation coefficient between each learning ability factor and the major of adult self-study examination, determine the target learning ability factor corresponding to the major of adult self-study examination from the multiple learning ability factors; for each test user, based on the target learning ability factor corresponding to the major of adult self-study examination to which the test user belongs, obtain the learning ability characteristic information of the test user.

[0010] Furthermore, based on the basic information, learning ability characteristic information, and self-study records of the multiple test users, classify the multiple test users to generate a classification result, including: grouping the multiple test users based on the adult self-study examination majors to which each of the test users belongs to generate multiple user units; for each of the user units, grouping the test users included in the user unit based on the regions to which the test users included in the user unit belong and the multiple region clusters corresponding to each of the adult self-study examination majors to generate multiple user groups included in the user unit; for each of the user groups, calculate the user similarity between any two test users included in the user group based on the learning ability characteristic information and self-study records of the multiple test users included in the user group, and cluster the multiple test users included in the user group according to the user similarity between any two test users included in the user group to determine multiple user clusters included in the user group, wherein the classification result includes the multiple user clusters included in each of the user groups.

[0011] Furthermore, based on the classification result, determine the multiple test users corresponding to each of the first adult self-study examination knowledge graphs, including: for each of the first adult self-study examination knowledge graphs, determine a first target user unit from the multiple user units based on the adult self-study examination major corresponding to the first adult self-study examination knowledge graph, determine a first target user group from the multiple user groups included in the first target user unit based on the region cluster corresponding to the first adult self-study examination knowledge graph, and use the test users included in the first target user group as the multiple test users corresponding to the first adult self-study examination knowledge graph; generate a second adult self-study examination knowledge graph corresponding to the first adult self-study examination knowledge graph based on the multiple test users corresponding to the first adult self-study examination knowledge graph, including: determining the association parameter between any two adult self-study examination knowledge points included in the first adult self-study examination knowledge graph based on the self-study records of the multiple test users corresponding to the first adult self-study examination knowledge graph; generating a second adult self-study examination knowledge graph corresponding to the first adult self-study examination knowledge graph based on the association parameter between any two adult self-study examination knowledge points included in the first adult self-study examination knowledge graph and the first adult self-study examination knowledge graph.

[0012] Further, based on the basic information, learning ability characteristic information of the user to be recommended, and the classification result, multiple reference test users corresponding to the user to be recommended and a second knowledge graph of adult self-study examination knowledge points are determined from the multiple test users, including: determining a second target user unit from the multiple user units based on the adult self-study examination major corresponding to the user to be recommended; determining a second target user group from the multiple user groups included in the second target user unit based on the region where the user to be recommended is located; determining a target user cluster from the user clusters included in the second target user group based on the basic information and learning ability characteristic information of the user to be recommended; determining multiple reference test users corresponding to the user to be recommended from the target user cluster based on the basic information and learning ability characteristic information of the user to be recommended; and determining a second knowledge graph of adult self-study examination knowledge points corresponding to the user to be recommended based on the adult self-study examination major and region corresponding to the user to be recommended.

[0013] Further, based on the self-study records of the user to be recommended, the corresponding multiple reference test users, and the second knowledge graph of adult self-study examination knowledge points, self-study examination recommendation content and recommended questions corresponding to the user to be recommended are generated, including: generating the self-study examination recommendation content and recommended questions corresponding to the user to be recommended through a recommendation prediction model based on the self-study records of the user to be recommended, the self-study records of the corresponding multiple reference test users, and the second knowledge graph of adult self-study examination knowledge points.

[0014] The present invention provides an intelligent adult self-study examination content and test question recommendation system for executing the above-mentioned intelligent adult self-study examination content and test question recommendation method, comprising: a test paper acquisition module for acquiring test papers of adult self-study examinations in previous years corresponding to different adult self-study examination majors in multiple regions; a map establishment module for establishing a first adult self-study examination knowledge point map corresponding to different adult self-study examination majors in each region based on test papers of adult self-study examinations in previous years corresponding to different adult self-study examination majors in multiple regions; a user testing module for acquiring basic information, learning ability characteristic information and self-study records of multiple test users; a user classification module for classifying the multiple test users based on the basic information, learning ability characteristic information and self-study records of the multiple test users to generate classification results; the map establishment module is also used to determine, based on the classification results, the multiple test users corresponding to each of the first adult self-study examination knowledge point maps. user; the map establishment module is also used to generate a second adult self-study examination knowledge point map corresponding to the first adult self-study examination knowledge point map for each of the first adult self-study examination knowledge point maps based on the multiple test users corresponding to the first adult self-study examination knowledge point map; the information acquisition module is used to obtain the basic information and learning ability characteristic information of the user to be recommended; the self-study examination recommendation module is used to determine the multiple reference test users and the second adult self-study examination knowledge point map corresponding to the user to be recommended from the multiple test users based on the basic information and learning ability characteristic information of the user to be recommended and the classification result; the information acquisition module is also used to obtain the self-study record of the user to be recommended; the self-study examination recommendation module is also used to generate the self-study examination recommended content and recommended test questions corresponding to the user to be recommended based on the self-study record of the user to be recommended, the corresponding multiple reference test users and the second adult self-study examination knowledge point map.

[0015] Compared with the existing technology, the intelligent adult self-study examination content and question recommendation system and method provided in this specification has at least the following beneficial effects:

[0016] 1. By analyzing the basic information, learning ability characteristic information, and self-study records of test users, personalized learning recommendations can be generated. This helps learners choose the most suitable learning resources according to their actual situations, improving learning efficiency. Based on the learning progress and knowledge point mastery of users, corresponding test questions are recommended, which helps learners consolidate knowledge and identify gaps. Through the analysis of past exam papers over the years, a detailed knowledge point map can be constructed, revealing the associations and dependencies between knowledge points. This helps educators better understand the curriculum structure, optimize teaching designs, and ensure the coherence and systematicness of teaching content. Personalized learning recommendations and targeted test question exercises help candidates better master knowledge points, improve problem-solving abilities and exam-taking skills, thus increasing the chances of passing the exam. Through precise learning resource recommendations, candidates can avoid blindly searching through a vast amount of learning materials and save precious learning time;

[0017] 2. By determining multiple learning ability factors and calculating their correlation coefficients with different adult self-study examination majors, a precise assessment of the learning ability of each test user can be achieved. This assessment not only considers the overall learning ability of users but also focuses on the ability factors closely related to specific majors. Based on the target learning ability factors, more personalized learning resources and strategies can be recommended for users, thereby improving learning effects and satisfaction. More reasonable adult self-study examination content and test questions can be recommended according to the learning ability characteristic information of users;

[0018] 3. By deeply analyzing adult self-study examination papers over the years, the exam knowledge points in each region and for each major can be comprehensively covered, ensuring the integrity and accuracy of the knowledge system. Not only are knowledge points extracted, but also characteristics such as the question proportion, question difficulty distribution, and question type distribution of each knowledge point are extracted, which helps in deeply understanding the exam requirements and trends. By calculating the overlap degree of exam knowledge points and the similarity of exam characteristics between regions, the similarity between regions can be accurately evaluated, providing a basis for clustering. For adult self-study examination majors in different regional clusters, a more targeted learning content and test question recommendation system can be constructed to meet the needs of learners in different regions and majors. Through personalized recommendations, learners can find suitable learning resources and paths faster, improving learning efficiency and effects. By constructing knowledge point maps and performing regional clustering, a large amount of valuable data resources have been accumulated, providing a scientific basis for adult self-study examination content and test question recommendations. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] This specification will be further described by way of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where:

[0020] Figure 1It is a flowchart of a method for recommending content and questions for intelligent adult self-study examinations shown in an embodiment of the present application;

[0021] Figure 2 It is a module diagram of a system for recommending content and questions for intelligent adult self-study examinations shown in an embodiment of the present application. Detailed implementation manners

[0022] To more clearly illustrate the technical solutions of the embodiments of this specification, the following will briefly introduce the accompanying drawings required for the description of the embodiments.

[0023] Figure 1 It is a flowchart of a method for recommending content and questions for intelligent adult self-study examinations shown in an embodiment of the present application. As Figure 1 shown, a method for recommending content and questions for intelligent adult self-study examinations may include the following steps.

[0024] Step 111: Obtain the previous-year adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions.

[0025] Specifically, the previous-year adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions can be obtained through web crawler technology from the official websites, databases or public materials of provincial education examination institutions, self-study examination offices or relevant educational institutions.

[0026] Step 112: Based on the previous-year adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions, establish a first adult self-study examination knowledge point map corresponding to each region for different adult self-study examination majors.

[0027] Specifically, it includes:

[0028] For each region, based on the previous-year adult self-study examination papers corresponding to different adult self-study examination majors in the region, determine the examination knowledge points corresponding to different adult self-study examination majors in the region and the examination characteristics corresponding to each examination knowledge point;

[0029] For each adult self-study examination major, based on the examination knowledge points corresponding to the adult self-study examination major in each region and the examination characteristics corresponding to each examination knowledge point, cluster multiple regions to determine multiple region clusters corresponding to the adult self-study examination major. Based on the examination knowledge points corresponding to the adult self-study examination major in each region cluster and the examination characteristics corresponding to each examination knowledge point, establish a first adult self-study examination knowledge point map corresponding to the region cluster for the adult self-study examination major.

[0030] Preferably, based on the previous-year adult self-study examination papers corresponding to different adult self-study examination majors in the region, determining the examination knowledge points corresponding to different adult self-study examination majors in the region and the examination characteristics corresponding to each examination knowledge point includes:

[0031] For each adult self-study examination major, a knowledge point database corresponding to the adult self-study examination major is constructed. For example, a knowledge point database corresponding to the adult self-study examination major can be constructed based on the examination syllabus, textbooks, or expert opinions;

[0032] For each adult self-study examination paper, preprocessing the adult self-study examination paper (for example, removing irrelevant information (such as question number, examinee name, etc.), formatting text (such as separating questions and answers), word segmentation, using optical character recognition technology to convert non-text data into text data, etc.) to generate a preprocessed adult self-study examination paper; and extracting knowledge points and knowledge point features from the preprocessed self-study examination paper based on a knowledge point library corresponding to an adult self-study examination major related to the adult self-study examination paper, wherein the knowledge point features at least include a question ratio, a question difficulty distribution, and a question type distribution for each knowledge point. The question ratio of a knowledge point can represent the ratio of the number of questions for the knowledge point included in the adult self-study examination paper to the total number of questions. The question difficulty distribution can represent the number of questions corresponding to different difficulty levels of the knowledge points included in the adult self-study examination paper. The question type distribution can represent the number of questions corresponding to different question types (for example, fill-in-the-blank questions, multiple-choice questions, short-answer questions, essay questions, etc.) of the knowledge points included in the adult self-study examination paper.

[0033] For each adult self-study examination major, based on the knowledge points and knowledge point characteristics of the adult self-study examination papers of the corresponding adult self-study examination major in the region, the examination knowledge points of the corresponding adult self-study examination major in the region and the examination characteristics corresponding to each examination knowledge point are determined.

[0034] Specifically, the knowledge points and knowledge point features of the adult self-study examination papers of the corresponding adult self-study examination majors in the region in previous years are weightedly fused to obtain the examination knowledge points of the corresponding adult self-study examination majors in the region and the examination features corresponding to each examination knowledge point, among which the examination features include the knowledge points of the corresponding adult self-study examination majors in the region and the proportion of test questions, test question difficulty distribution and question type distribution for each knowledge point.

[0035] Preferably, based on the examination knowledge points of the adult self-study examination major corresponding to each region and the examination features corresponding to each examination knowledge point, multiple regions are clustered to determine multiple regional clusters corresponding to the adult self-study examination major;

[0036] Based on the examination knowledge points of the adult self-study examination major corresponding to each region, calculate the overlap of the examination knowledge points of the adult self-study examination major corresponding to any two regions;

[0037] Based on the examination features corresponding to each examination knowledge point of the adult self-study examination major in each region, calculate the similarity of the examination features of the adult self-study examination majors in any two regions;

[0038] Calculate the regional similarity of the adult self-study examination majors corresponding to any two regions based on the overlap degree of the examination knowledge points of the adult self-study examination majors corresponding to any two regions and the similarity of the examination characteristics of the adult self-study examination majors corresponding to any two regions.

[0039] Cluster multiple regions based on the regional similarity of the adult self-study examination majors corresponding to any two regions, and determine multiple regional clusters corresponding to the adult self-study examination majors.

[0040] Specifically, the similarity of the examination characteristics of the adult self-study examination majors corresponding to two regions can be calculated according to the following formula:

[0041] S ((i，j)，k，1) = a 11 × cos(F ((i，k，1) , F ((j，k，1) ) + a 12 × cos(F ((i，k，2) , F ((j，k，2) ) + a 13 × cos(F ((i，k，3) , F ((j，k，3) )

[0042] Where S ((i，j)，k，1) is the similarity of the examination characteristics of the k-th adult self-study examination major corresponding to the i-th region and the j-th region, a 11 , a 12 and a 13 are preset weights, a 11 , a 12 and a 13 are greater than 0, a 11 + a 12 + a 13 = 1, cos(F ((i，k，1) , F ((j，k，1) ) is the cosine similarity between the question proportion of each knowledge point of the k-th adult self-study examination major corresponding to the i-th region and the question proportion of each knowledge point of the k-th adult self-study examination major corresponding to the j-th region, cos(F ((i，k，2) , F ((j，k，2) ) is the cosine similarity between the question difficulty distribution of each knowledge point of the k-th adult self-study examination major corresponding to the i-th region and the question difficulty distribution of each knowledge point of the k-th adult self-study examination major corresponding to the j-th region, cos(F ((i，k，3) , F ((j，k，3) ) is the cosine similarity between the question type distribution of each knowledge point of the k-th adult self-study examination major corresponding to the i-th region and the question type distribution of each knowledge point of the k-th adult self-study examination major corresponding to the j-th region.

[0043] The regional similarity of any two regions corresponding to the majors of adult self-study examinations is as follows according to the following formula:

[0044] S ((i，j)，k，2) =b 11 ×C ((i，j)，k) +b 12 ×S ((i，j)，k，1)

[0045] Among them, S ((i，j)，k，2) is the regional similarity of the i-th region and the j-th region corresponding to the k-th major of the adult self-study examination, b 11 and b 12 are preset weights, b 11 and b 12 are greater than 0, b 11 +b 12 =1, C ((i，j)，k) is the coincidence degree of examination knowledge points of the i-th region and the j-th region corresponding to the k-th major of the adult self-study examination.

[0046] Based on the regional similarity of any two regions corresponding to the majors of the adult self-study examination, multiple regions are clustered by the K-means clustering algorithm to determine multiple region clusters corresponding to the majors of the adult self-study examination.

[0047] The first adult self-study examination knowledge point graph may include nodes representing multiple knowledge points corresponding to different majors of the adult self-study examination in the region.

[0048] Step 113, obtain the basic information, learning ability characteristic information and self-study records of multiple test users.

[0049] Only as an example, the basic information of the test user may at least include age, occupation, education level, adult self-study goals (such as region, major of the adult self-study examination, target school, etc.). The learning ability characteristic information may at least include cognitive style, learning preference, attention concentration, etc. The self-study record may include the knowledge points that have been completed in learning, learning progress, practice completion situation and learning duration of the test user at multiple self-study time points.

[0050] Step 114, classify multiple test users based on the basic information, learning ability characteristic information and self-study records of multiple test users, and generate a classification result.

[0051] Specifically including:

[0052] Group multiple test users based on the majors of the adult self-study examination to which each test user belongs, and generate multiple user units, where one major of the adult self-study examination corresponds to one user unit;

[0053] For each user unit, based on the region to which the test users included in the user unit belong and the multiple region clusters corresponding to each adult self-study examination major, group the test users included in the user unit to generate multiple user groups included in the user unit. Among them, one user group included in the user unit corresponds to one region cluster corresponding to the adult self-study examination major to which the user unit belongs;

[0054] For each user group, based on the learning ability characteristic information and self-study records of the multiple test users included in the user group, calculate the user similarity between any two test users included in any two user groups. According to the user similarity between any two test users included in any two user groups, cluster the multiple test users included in the user group to determine multiple user clusters included in the user group. Among them, the classification result includes multiple user clusters included in each user group.

[0055] Specifically, a similarity determination model can be used to calculate the user similarity between any two test users included in any two user groups based on the learning ability characteristic information and self-study records of the multiple test users included in the user group. Among them, the similarity determination model can be a convolutional neural network model. Use the K-means clustering algorithm to cluster the multiple test users included in the user group according to the user similarity between any two test users included in any two user groups to determine multiple user clusters included in the user group.

[0056] Step 115, based on the classification result, determine multiple test users corresponding to each first adult self-study examination knowledge graph.

[0057] Specifically, it includes:

[0058] For each first adult self-study examination knowledge graph, based on the adult self-study examination major corresponding to the first adult self-study examination knowledge graph, determine a first target user unit from multiple user units. Based on the region cluster corresponding to the first adult self-study examination knowledge graph, determine a first target user group from the multiple user groups included in the first target user unit. Use the test users included in the first target user group as the multiple test users corresponding to the first adult self-study examination knowledge graph.

[0059] For example, if the adult self-study examination major corresponding to the first adult self-study examination knowledge graph is A, then use the user unit corresponding to the adult self-study examination major A as the first target user unit. The first target unit includes user groups B1, B2, and B3. User group B1 corresponds to region cluster C1, user group B2 corresponds to region cluster C2, and user group B3 corresponds to region cluster C3. And if the first adult self-study examination knowledge graph corresponds to region cluster C1, then use user group B1 as the first target user group.

[0060] Step 116: For each first adult self-study examination knowledge point graph, generate a second adult self-study examination knowledge point graph corresponding to the first adult self-study examination knowledge point graph based on multiple test users corresponding to the first adult self-study examination knowledge point graph.

[0061] Specifically, it includes:

[0062] Determine the correlation parameter between any two adult self-study examination knowledge points included in the first adult self-study examination knowledge point graph based on the self-study records of multiple test users corresponding to the first adult self-study examination knowledge point graph;

[0063] Generate a second adult self-study examination knowledge point graph corresponding to the first adult self-study examination knowledge point graph based on the correlation parameter between any two adult self-study examination knowledge points included in the first adult self-study examination knowledge point graph and the first adult self-study examination knowledge point graph.

[0064] Specifically, for any two adult self-study examination knowledge points (for example, adult self-study examination knowledge point e and adult self-study examination knowledge point f), according to the self-study records of multiple test users corresponding to the first adult self-study examination knowledge point graph, the multiple test users corresponding to the first adult self-study examination knowledge point graph can be divided into three categories. Among them, the first category is the test users who have only studied adult self-study examination knowledge point e, the second category is the test users who have only studied adult self-study examination knowledge point f and the test users who have completed the study of both adult self-study examination knowledge point e and adult self-study examination knowledge point f. According to the test scores of the first category of test users corresponding to adult self-study examination knowledge point e, the test scores of the second category of test users corresponding to adult self-study examination knowledge point f, and the test scores of the third category of test users corresponding to both adult self-study examination knowledge point e and adult self-study examination knowledge point f, calculate the correlation parameter between the two adult self-study examination knowledge points.

[0065] The correlation parameter between the two adult self-study examination knowledge points can be calculated according to the following formula:

[0066] C (e，f)) =c 11 ×C (e，f)，1) +c 12 ×C (e，f)，2)

[0067]

[0068]

[0069] Among them, C (e，f)) is the correlation parameter between the e-th adult self-study examination knowledge point and the f-th adult self-study examination knowledge point, c 11 and c 12 are preset weights, c11 and c 12 is greater than 0, c 11 +c 12 = 1, C (e，f)，1) is the first single-point correlation parameter between the e-th self-study examination knowledge point for adults and the f-th self-study examination knowledge point for adults, C (e，f)，2) is the second single-point correlation parameter between the e-th self-study examination knowledge point for adults and the f-th self-study examination knowledge point for adults, is the test score of the m1-th test user of the first category corresponding to the self-study examination knowledge point e for adults, M1 is the total number of test users of the first category, is the test score of the m3-th test user of the third category corresponding to the self-study examination knowledge point e for adults, M3 is the total number of test users of the third category, M2 is the total number of test users of the second category, is the test score of the m2-th test user of the second category corresponding to the self-study examination knowledge point f for adults, is the test score of the m3-th test user of the third category corresponding to the self-study examination knowledge point f for adults.

[0070] The second self-study examination knowledge point map for adults may include nodes representing multiple knowledge points corresponding to different self-study examination majors for adults and edges connecting between the nodes corresponding to the two self-study examination knowledge points for adults and used to represent the correlation parameters between the two self-study examination knowledge points. The shorter the edge, the greater the correlation parameter between the two self-study examination knowledge points.

[0071] Step 117, obtain the basic information and learning ability characteristic information of the user to be recommended.

[0072] Specifically include:

[0073] Determine multiple learning ability factors;

[0074] Obtain the ability test data of multiple test users. Among them, the ability test data of the test users may include the scores of the test users in each learning ability factor;

[0075] For each self-study examination major for adults, based on the ability test data of multiple test users, determine the correlation coefficient between each learning ability factor and the self-study examination major for adults. Based on the correlation coefficient between each learning ability factor and the self-study examination major for adults, determine the target learning ability factor corresponding to the self-study examination major from multiple learning ability factors;

[0076] For each test user, based on the target learning ability factor corresponding to the self-study examination major to which the test user belongs, obtain the learning ability characteristic information of the test user.

[0077] Specifically, for each learning ability factor, the correlation coefficient between the learning ability factor and the adult self-study examination major can be calculated based on the self-study efficiency of multiple test users belonging to the adult self-study examination major and their scores in this learning ability factor.

[0078] For example, the correlation coefficient between the learning ability factor and the adult self-study examination major can be calculated according to the following formula:

[0079]

[0080] where r (P，Q) is the correlation coefficient between the P-th learning ability factor and the Q-th adult self-study examination major, M4 is the total number of test users belonging to the adult self-study examination major, is the score of the m4-th test user belonging to the adult self-study examination major in the P-th learning ability factor, is the self-study efficiency of the m4-th test user belonging to the adult self-study examination major in the Q-th adult self-study examination major.

[0081] Step 118: Based on the basic information, learning ability characteristic information, and classification result of the user to be recommended, determine multiple reference test users and the second adult self-study examination knowledge graph corresponding to the user to be recommended from multiple test users.

[0082] Specifically, it includes:

[0083] Based on the adult self-study examination major corresponding to the user to be recommended, determine the second target user unit from multiple user units;

[0084] Based on the region where the user to be recommended is located, determine the second target user group from multiple user groups included in the second target user unit;

[0085] Based on the basic information and learning ability characteristic information of the user to be recommended, determine the target user group from the user clusters included in the second target user group;

[0086] Based on the basic information and learning ability characteristic information of the user to be recommended, determine multiple reference test users corresponding to the user to be recommended from the target user group;

[0087] Based on the adult self-study examination major and the region where the user to be recommended is located, determine the second adult self-study examination knowledge graph corresponding to the user to be recommended.

[0088] Step 119: Obtain the self-study record of the user to be recommended.

[0089] Step 120: Based on the self-study record of the user to be recommended, the corresponding multiple reference test users, and the second adult self-study examination knowledge graph, generate the self-study examination recommendation content and recommended questions corresponding to the user to be recommended.

[0090] Specifically include:

[0091] Through the recommendation prediction model, based on the self-study records of the user to be recommended, the self-study records of multiple corresponding reference test users, and the second adult self-study examination knowledge graph, generate the self-study examination recommendation content and recommended questions corresponding to the user to be recommended, where the recommendation prediction model can be a convolutional neural network model.

[0092] Figure 2 It is a module diagram of an intelligent adult self-study examination content and question recommendation system shown in an embodiment of the present application. As Figure 2 shown, an intelligent adult self-study examination content and question recommendation system may include a test paper acquisition module, a knowledge graph establishment module, a user test module, a user classification module, an information acquisition module, and a self-study examination recommendation module.

[0093] The test paper acquisition module can be used to acquire the previous adult self-study examination test papers corresponding to different adult self-study examination majors in multiple regions;

[0094] The knowledge graph establishment module can be used to establish the first adult self-study examination knowledge graph corresponding to each region and different adult self-study examination majors based on the previous adult self-study examination test papers corresponding to different adult self-study examination majors in multiple regions;

[0095] The user test module can be used to acquire the basic information, learning ability characteristic information, and self-study records of multiple test users;

[0096] The user classification module can be used to classify multiple test users based on the basic information, learning ability characteristic information, and self-study records of multiple test users, and generate a classification result;

[0097] The knowledge graph establishment module is also used to determine multiple test users corresponding to each first adult self-study examination knowledge graph based on the classification result;

[0098] The knowledge graph establishment module is also used to generate the second adult self-study examination knowledge graph corresponding to each first adult self-study examination knowledge graph based on multiple test users corresponding to the first adult self-study examination knowledge graph for each first adult self-study examination knowledge graph;

[0099] The information acquisition module can be used to acquire the basic information and learning ability characteristic information of the user to be recommended;

[0100] The self-study examination recommendation module can be used to determine multiple reference test users and the second adult self-study examination knowledge graph corresponding to the user to be recommended from multiple test users based on the basic information, learning ability characteristic information, and classification result of the user to be recommended;

[0101] The information acquisition module is further configured to acquire the self-study records of the user to be recommended;

[0102] The self-study exam recommendation module is further configured to generate self-study exam recommendation content and recommended questions corresponding to the user to be recommended based on the self-study records of the user to be recommended, the corresponding multiple reference test users, and the second adult self-study exam knowledge graph.

[0103] An intelligent adult self-study exam content and question recommendation system can be used to execute an intelligent adult self-study exam content and question recommendation method. For more descriptions of an intelligent adult self-study exam content and question recommendation system, reference can be made to the relevant descriptions of an intelligent adult self-study exam content and question recommendation method, which will not be elaborated here.

[0104] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification can be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. An intelligent method for recommending the content and questions of the adult self-study examination, characterized in that, Comprising: Obtaining the past adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions; Based on the past adult self-study examination papers corresponding to different adult self-study examination majors in multiple regions, establishing the first adult self-study examination knowledge point map corresponding to each region for different adult self-study examination majors; Obtaining the basic information, learning ability characteristic information and self-study records of multiple test users; Based on the basic information, learning ability characteristic information and self-study records of the multiple test users, classifying the multiple test users to generate a classification result; Based on the classification result, determining multiple test users corresponding to each of the first adult self-study examination knowledge point maps; For each of the first adult self-study examination knowledge point maps, based on the multiple test users corresponding to the first adult self-study examination knowledge point map, generating the second adult self-study examination knowledge point map corresponding to the first adult self-study examination knowledge point map; Obtaining the basic information and learning ability characteristic information of the user to be recommended; Based on the basic information and learning ability characteristic information of the user to be recommended and the classification result, determining multiple reference test users and the second adult self-study examination knowledge point map corresponding to the user to be recommended from the multiple test users; Obtaining the self-study record of the user to be recommended; Based on the self-study record of the user to be recommended, the corresponding multiple reference test users and the second adult self-study examination knowledge point map, generating the self-study examination recommendation content and recommended questions corresponding to the user to be recommended. Specifically, through a recommendation prediction model, based on the self-study record of the user to be recommended, the self-study records of the corresponding multiple reference test users and the second adult self-study examination knowledge point map, generating the self-study examination recommendation content and recommended questions corresponding to the user to be recommended; Based on the basic information, learning ability characteristic information and self-study records of the multiple test users, classifying the multiple test users to generate a classification result, including: Grouping the multiple test users based on the adult self-study examination major to which each test user belongs to generate multiple user units; For each of the user units, grouping the test users included in the user unit based on the region to which the test users included in the user unit belong and the multiple region clusters corresponding to each adult self-study examination major to generate multiple user groups included in the user unit; For each of the user groups, calculating the user similarity between any two test users included in the user group based on the learning ability characteristic information and self-study records of the multiple test users included in the user group, and clustering the multiple test users included in the user group according to the user similarity between any two test users included in the user group to determine multiple user clusters included in the user group, wherein the classification result includes the multiple user clusters included in each of the user groups; Based on the classification result, determining multiple test users corresponding to each of the first adult self-study examination knowledge point maps, including: For each of the first self-study exam knowledge graphs for adults, based on the adult self-study exam majors corresponding to the first self-study exam knowledge graphs for adults, determine a first target user unit from the multiple user units. Based on the regional clusters corresponding to the first self-study exam knowledge graphs for adults, determine a first target user group from the multiple user groups included in the first target user unit. Use the test users included in the first target user group as the multiple test users corresponding to the first self-study exam knowledge graphs for adults; Generate a second self-study exam knowledge graph for adults corresponding to the first self-study exam knowledge graph for adults based on the multiple test users corresponding to the first self-study exam knowledge graph for adults, including: Determine the association parameters between any two self-study exam knowledge points for adults included in the first self-study exam knowledge graph for adults based on the self-study records of the multiple test users corresponding to the first self-study exam knowledge graph for adults; Generate a second self-study exam knowledge graph for adults corresponding to the first self-study exam knowledge graph for adults based on the association parameters between any two self-study exam knowledge points for adults included in the first self-study exam knowledge graph for adults and the first self-study exam knowledge graph for adults.

2. The intelligent adult self-study examination content and test question recommendation method according to claim 1, wherein Establish a first self-study exam knowledge graph for adults corresponding to each region for different adult self-study exam majors based on the past years' adult self-study exam papers corresponding to different adult self-study exam majors in multiple regions, including: For each region, based on the past years' adult self-study exam papers corresponding to different adult self-study exam majors in the region, determine the exam knowledge points corresponding to different adult self-study exam majors in the region and the exam characteristics corresponding to each exam knowledge point; For each adult self-study exam major, cluster the multiple regions based on the exam knowledge points corresponding to the adult self-study exam major in each region and the exam characteristics corresponding to each exam knowledge point, determine multiple regional clusters corresponding to the adult self-study exam major, and establish a first self-study exam knowledge graph for adults corresponding to the regional cluster for the adult self-study exam major based on the exam knowledge points corresponding to the adult self-study exam major in the multiple regions included in each regional cluster and the exam characteristics corresponding to each exam knowledge point.

3. An intelligent self-study examination content and question recommendation method for adults according to claim 2, characterized in that, Determine the exam knowledge points corresponding to different adult self-study exam majors in the region and the exam characteristics corresponding to each exam knowledge point based on the past years' adult self-study exam papers corresponding to different adult self-study exam majors in the region, including: For each of the adult self-study exam majors, construct a knowledge point library corresponding to the adult self-study exam major; For each adult self-study exam paper, preprocess the adult self-study exam paper to generate a preprocessed adult self-study exam paper, and extract knowledge points and knowledge point characteristics from the preprocessed self-study exam paper according to the knowledge point library corresponding to the adult self-study exam major related to the adult self-study exam paper, where the knowledge point characteristics at least include the question proportion, question difficulty distribution, and question type distribution of each knowledge point; For each of the adult self-study examination majors, based on the knowledge points and knowledge point features of the adult self-study examination papers of previous years corresponding to the adult self-study examination major in the area, the examination knowledge points corresponding to the adult self-study examination major in the area and the examination features corresponding to each examination knowledge point are determined.

4. The intelligent adult self-study examination content and test question recommendation method according to claim 2, wherein, Clustering the multiple regions based on the examination knowledge points corresponding to the adult self-study examination major in each region and the examination features corresponding to each examination knowledge point to determine multiple regional clusters corresponding to the adult self-study examination major; Based on the examination knowledge points of the adult self-study examination major corresponding to each of the areas, calculating the overlap of the examination knowledge points of the adult self-study examination major corresponding to any two areas; Based on the examination features corresponding to each examination knowledge point of the adult self-study examination major corresponding to each of the regions, calculating the similarity of the examination features corresponding to the adult self-study examination majors of any two regions; Calculate the regional similarity between any two regions corresponding to the adult self-study examination majors based on the overlap of examination knowledge points between any two regions corresponding to the adult self-study examination majors and the similarity of examination features between any two regions corresponding to the adult self-study examination majors; Based on the regional similarity between any two of the regions corresponding to the adult self-study examination majors, the multiple regions are clustered to determine multiple regional clusters corresponding to the adult self-study examination majors.

5. A method for recommending intelligent adult self-study examination content and test questions according to any one of claims 1-4, characterized in that, Obtain the learning ability characteristic information of the test user, including: Identify multiple learning ability factors; Obtain ability test data of multiple test users; For each of the adult self-study examination majors, determining a correlation coefficient between each learning ability factor and the adult self-study examination major based on the ability test data of the multiple test users, and determining a target learning ability factor corresponding to the adult self-study examination major from the multiple learning ability factors based on the correlation coefficient between each learning ability factor and the adult self-study examination major; For each of the test users, the learning ability characteristic information of the test user is obtained based on the target learning ability factor corresponding to the adult self-study examination major to which the test user belongs.

6. The intelligent content and question recommendation method for adult self-study examinations according to claim 1, characterized in that Based on the basic information and learning ability characteristic information of the user to be recommended and the classification result, determining multiple reference test users and a second adult self-study examination knowledge point map corresponding to the user to be recommended from the multiple test users, including: Determining a second target user unit from the plurality of user units based on the adult self-study examination major corresponding to the user to be recommended; Determining a second target user group from a plurality of user groups included in the second target user unit based on the area to which the to-be-recommended user belongs; Determining a target user cluster from the user clusters included in the second target user group based on the basic information and learning ability characteristic information of the user to be recommended; Based on the basic information and learning ability characteristic information of the user to be recommended, determining a plurality of reference test users corresponding to the user to be recommended from the target user cluster; Based on the adult self-study examination major and the region to which the to-be-recommended user corresponds, a second adult self-study examination knowledge point map corresponding to the to-be-recommended user is determined.

7. An intelligent content and question recommendation system for adult self-study examinations, characterized in that, A method for executing an intelligent adult self-study examination content and question recommendation method as described in any one of claims 1-6, including: A test paper acquisition module for acquiring previous-year adult self-study examination test papers corresponding to different adult self-study examination majors in multiple regions; A knowledge graph building module for building a first adult self-study examination knowledge graph corresponding to each region and different adult self-study examination majors based on the previous-year adult self-study examination test papers corresponding to different adult self-study examination majors in multiple regions; A user test module for acquiring the basic information, learning ability characteristic information, and self-study records of multiple test users; A user classification module for classifying the multiple test users based on the basic information, learning ability characteristic information, and self-study records of the multiple test users to generate a classification result; The knowledge graph building module is further configured to determine multiple test users corresponding to each of the first adult self-study examination knowledge graphs based on the classification result; The knowledge graph building module is further configured to generate a second adult self-study examination knowledge graph corresponding to each of the first adult self-study examination knowledge graphs based on the multiple test users corresponding to the first adult self-study examination knowledge graph; An information acquisition module for acquiring the basic information and learning ability characteristic information of a user to be recommended; A self-study examination recommendation module for determining multiple reference test users and a second adult self-study examination knowledge graph corresponding to the user to be recommended from the multiple test users based on the basic information and learning ability characteristic information of the user to be recommended and the classification result; The information acquisition module is further configured to acquire the self-study record of the user to be recommended; The self-study examination recommendation module is further configured to generate self-study examination recommendation content and recommended questions corresponding to the user to be recommended based on the self-study record of the user to be recommended, the corresponding multiple reference test users, and the second adult self-study examination knowledge graph. Specifically, through a recommendation prediction model, based on the self-study record of the user to be recommended, the self-study records of the corresponding multiple reference test users, and the second adult self-study examination knowledge graph, generate the self-study examination recommendation content and recommended questions corresponding to the user to be recommended.

Images