Self-adaptive learning system and method based on modular question bank

Through the adaptive learning system of modular question bank and knowledge tree algorithm, the problems of insufficient self-control and insufficient guidance in traditional education are solved, and personalized questions push and review strategies are realized, learning efficiency and knowledge mastery are improved, and students are enhanced. initiative and interest in learning.

CN119090682APending Publication Date: 2024-12-06郑尘进
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410532689.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-04-30
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

In the traditional education system, students lack self-control, cannot obtain timely guidance, and find it difficult to accurately locate blind spots of knowledge and arrange review plans independently, resulting in low learning efficiency and forgetfulness.

Method used

Design an adaptive learning system based on a modular question bank, including the question bank module, learning module and review module, split the test questions through the tree diagram structure, dynamically push questions and analysis, use algorithms to push review questions, and introduce incentive modules to provide virtual award.

Benefits of technology

It improves students' learning autonomy and efficiency, provides personalized guidance, helps students quickly identify and supplement knowledge blind spots, extends the time to maintain knowledge, and enhances learning interest and motivation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure FT_1
    Figure FT_1
  • Figure FT_2
    Figure FT_2
  • Figure FT_3
    Figure FT_3
Patent Text Reader

Abstract

The invention discloses a self-adaptive learning system and method based on a modular question bank, which are used for solving the problems that students are insufficient in self-control, cannot obtain timely guidance and the like, and improving the learning effect. The system comprises a question bank module, a learning module, a review module and an excitation module. The test questions in the question bank module are split into one or more sub-questions or knowledge points to form a tree diagram structure. And the learning module dynamically pushes questions and carries out detailed analysis according to the answering conditions of the students. The review module pushes the questions again at the critical point where the students are about to forget knowledge, and preferentially pushes the questions with the highest complexity. And the incentive module gives virtual rewards according to the number of knowledge points mastered by the students. According to the invention, a personalized question pushing and reviewing strategy for each student is realized by using a modular question bank and a knowledge tree algorithm. Real-time feedback and detailed analysis in the learning process can help students to quickly check leakage and fill vacancies, and a review mechanism based on a forgetting curve prolongs the retention time of knowledge to the greatest extent. And meanwhile, the initiative and the participation degree of the students are also improved through a gamified reward and feedback mechanism. Compared with a traditional learning mode, the method can remarkably improve the learning efficiency and knowledge mastering degree of students, and is of great significance to improvement of education quality.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of online education technology, and in particular to an adaptive learning system and method based on a modular question bank, which are used to solve the problems of students' lack of self-control and inability to obtain timely guidance, and improve learning effects. Background Art

[0002] In the traditional education system, there are the following problems: (1) Some students lack self-control and find it difficult to study independently without teacher supervision; (2) Students cannot get timely and personalized guidance when they encounter difficult problems, and cannot get guidance before they forget the questions. They are easily frustrated and resist because they cannot solve the questions; (3) Students are inefficient in looking up answers on their own and find it difficult to accurately locate knowledge blind spots. (4) Students cannot arrange a reasonable review plan on their own and are prone to forget. Summary of the invention

[0003] In order to solve the above technical problems, the present invention proposes an adaptive learning system and method based on a modular question bank. The system includes a question bank module, a learning module, a review module and an incentive module.

[0004] The test questions in the question bank module are divided into one or more sub-questions or knowledge points, forming a tree structure. The learning module dynamically pushes questions and detailed explanations based on the students' answers. The review module pushes questions again when students are about to forget the knowledge, giving priority to the most complex questions. The incentive module gives virtual rewards based on the number of knowledge points students have mastered.

[0005] Compared with the prior art, the present invention has the following beneficial effects: (1) the gamified interaction and reward mechanism helps to improve students' learning autonomy; (2) the detailed question analysis can replace the teacher's one-on-one guidance; (3) the modular question bank design and review strategy enable students to clearly understand the question structure and accurately identify blind spots; (4) the review of questions pushed according to the algorithm can reduce students' forgetting. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 This is a schematic diagram of the system architecture of the present invention, showing the interactive relationship between various functional modules.

[0007] Figure 2 This is a data structure diagram of the question bank module, showing the properties and tree topology of the question nodes.

[0008] Figure 3 This is an example flow chart of the learning module combined with the question bank module, showing the process of dynamically pushing questions based on student feedback. DETAILED DESCRIPTION

[0009] The core of the present invention is the design and implementation of a modular question bank. Each question in the question bank module is split into one or more sub-questions or knowledge points and organized into a tree-like data structure. Each question node contains the following attributes:

[0010] - Title: The main content and analysis of the title.

[0011] - Complexity: The level of the question in the tree diagram reflects the complexity of the question.

[0012] - Parent node pointer: points to the parent node of the question.

[0013] Subnode pointer array: points to the next level node of the question.

[0014] It should be noted that the pointers between question nodes do not have a clear direction, but only indicate the relationship between nodes. The complexity of a question is determined by its level in the knowledge tree, which is an objective attribute. However, considering that the complexity of a question can be directly calculated by traversing the level information of the question nodes, it is not necessary to store it as a separate field when designing the database. This can simplify the data structure and improve the efficiency of storage and retrieval. At the same time, this design also provides greater flexibility for subsequent algorithm adjustment and optimization.

[0015] The core algorithm of the learning module is as follows:

[0016] 1. The system selects any question from the question bank and pushes it to the student. In best practice, leaf node questions with the lowest complexity should be selected first.

[0017] 2. If the student gets this question right:

[0018] - Mark this question as solved.

[0019] - Check if this question has any sibling nodes (other sub-questions in the same branch):

[0020] - If yes, push the next sibling node topic;

[0021] - If not, push the parent node title.

[0022] 3. If the student gets the question wrong or selects "No":

[0023] - The system first pushes a detailed analysis of the question to help students understand the question;

[0024] - Then, the system checks whether this question has child nodes (questions at the next level):

[0025] - If yes, push the first child node title;

[0026] - If not, push the current question again to give students another chance to try.

[0027] 4. When the student has completed all the questions in the current branch, the system can push a new question branch, giving priority to the leaf node questions with the lowest complexity in the branch.

[0028] 5. Repeat steps 2-4 until students have completed all the questions.

[0029] The review module tracks the students' answers and time on each question, uses an algorithm to predict when students may forget a knowledge point, and then re-sends the question review before the critical point. The input of the algorithm is the students' answers and timestamps on each question, and the output is the estimated forgetting time point for each knowledge point.

[0030] For a single question, the main rules of the algorithm are as follows (the specific data in the actual situation may be slightly different):

[0031] - If the student answers correctly, the next review period will be twice the current review period;

[0032] - If the student answers incorrectly, the next review time will be equal to the current review cycle time;

[0033] - For the first review, the next review will be 1 day later;

[0034] - If the calculated next review time is less than 1 day, it will be set to 1 day later.

[0035] When the review module pushes review questions, it gives priority to the questions with the highest complexity in the knowledge tree where the question is located. If the student can answer it correctly, the review is considered successful, and the question and all the questions in its subordinate branches are marked as successful.

[0036] If the student answers incorrectly, the knowledge tree where the question is located will be reactivated to recursively review the lower-level questions. Specifically, the system will push the sub-node questions of the current question for students to re-learn and practice. If the current question has no sub-nodes, the question will be pushed again for review.

[0037] The incentive module is responsible for giving virtual rewards such as coins, points, and titles based on the number of knowledge points that students have mastered. These rewards can be used to unlock new learning content or exchange for physical prizes to increase students' learning motivation. In addition, the system can also provide personalized encouragement and feedback based on students' learning progress and performance to help students maintain their interest in learning.

[0038] In summary, the adaptive learning system provided by the present invention utilizes a modular question bank and a knowledge tree algorithm to implement personalized question push and review strategies for each student. Real-time feedback and detailed analysis during the learning process can help students quickly find and fill gaps, while the review mechanism based on the forgetting curve maximizes the retention time of knowledge. At the same time, the gamification reward and feedback mechanism also improves students' initiative and participation. Compared with traditional learning methods, the present invention can significantly improve students' learning efficiency and knowledge mastery, which is of great significance to improving the quality of education.

Claims

1. An adaptive learning system based on a modular question bank, characterized by , including question bank module, learning module, review module and incentive module; the test questions in the question bank module are divided into knowledge points and form a tree structure; the learning module pushes questions and solutions according to the students' answers; the review module predicts the time when students forget knowledge based on the forgetting curve and pushes review questions; the incentive module gives virtual rewards according to the number of knowledge points mastered by students.

2. The system according to claim 1, characterized in that ,Each question node in the question bank module contains the question stem, complexity, parent node pointer and child node ,pointer array.

3. The system according to claim 1, characterized in that ,The algorithm steps for the learning module to push questions are as follows: Select any question from the question bank, giving priority to the leaf node question with the lowest complexity; If the student gets it right, the current question is marked as mastered, and the sibling node question or parent node question is pushed; If students make mistakes, detailed explanations and sub-node questions or original questions will be pushed for further practice; When students complete the current branch, they can push questions from a new branch; repeat the above steps until students complete all questions.

4. The system according to claim 1, characterized in that ,The push strategies for the review module include: Use the forgetting curve algorithm to predict when students are likely to forget knowledge points; Before the critical point of forgetting, the most complex questions in the branch where the questions are located are pushed first; If the student answers correctly, the review is considered successful; if the answer is incorrect, the knowledge tree where the question is located is reactivated and the lower-level questions are reviewed recursively.

5. The system according to claim 1, characterized in that ,The incentive module gives virtual rewards such as gold coins, points, and titles based on the ,number of knowledge points mastered by students, which can be used to unlock new ,content or redeem prizes.

6. An adaptive learning method based on the system according to any one of claims 1 to 5, characterized in that , including the following steps: Select questions from the modular question bank, giving priority to leaf node questions with the lowest complexity; If the student gets it right, the current question will be marked as mastered, and the sibling node question or parent node question will be pushed; if the student gets it wrong, the detailed explanation and the child node question or the original question will be pushed for further practice; When students complete the current branch, they can push questions for the new branch; Using the forgetting curve algorithm, we prioritize reviewing the most complex questions in the branch where the questions are located before students reach the critical point of forgetting the knowledge. Give virtual rewards based on the number of knowledge points students have mastered; Repeat the above steps until students have completed all the questions.