Knowledge point learning recommendation method and system based on course video

Abstract

The invention discloses a knowledge point learning recommendation method and system based on a course video, belongs to the technical field of intelligent education, and can solve the problem that an existing course video learning mode is difficult to meet personalized learning requirements of students. The method comprises the following steps: S1, determining an association relationship among a plurality of knowledge points, and processing a course video by using an optical character recognition method and a large language model based on the plurality of knowledge points to generate a learning video of each knowledge point; s2, determining knowledge points in each test question in the question bank by using a large language model, and determining a learning corresponding relationship among the knowledge points, the learning video and the test questions; and S3, constructing a knowledge graph according to the association relationship and the learning corresponding relationship, and pushing a learning video and test questions matched with a specific demand to the knowledge demander according to the specific demand of the knowledge demander and the knowledge graph. The method is used for knowledge point learning recommendation.

Description

Technical Field

[0001] The present invention relates to a method and system for recommending knowledge point learning based on course videos, belonging to the technical field of smart education. Background Art

[0002] In the field of medical and nursing learning, teachers generally record multiple course videos through live broadcast, recorded broadcast, face-to-face teaching, etc., and students then learn the course by watching the course videos in sequence. Since the course videos usually correspond to the chapters of the textbook, each course video is relatively long and contains many knowledge points, which is more suitable for students with no basic knowledge or sufficient study time to conduct systematic learning.

[0003] However, for students who have a certain foundation and limited study time, they do not need to learn all the knowledge points one by one, but only need to strengthen their study of the knowledge points that they have not mastered. However, the existing course videos have a large number of knowledge points, a long learning time, and a single learning method. They cannot support students to quickly and accurately find the knowledge points that they have not mastered and conduct targeted learning, so it is difficult to meet the students' personalized learning needs. This not only increases the students' learning burden and reduces their learning enthusiasm, but also seriously affects their learning efficiency and learning results. Summary of the invention

[0004] The present invention provides a method and system for recommending knowledge points based on course videos, which can solve the problem that the existing course video learning method is difficult to meet the personalized learning needs of students.

[0005] In one aspect, the present invention provides a method for recommending knowledge points based on course videos, the method comprising: S1. Determine the correlation between multiple knowledge points, and based on the multiple knowledge points, use the optical character recognition method and the large language model to process the course video to generate a learning video for each knowledge point; S2. Use the large language model to determine the knowledge points in each test question in the question bank, and determine the learning correspondence between the knowledge points, learning videos, and test questions; S3. Construct a knowledge graph based on the association relationship and the learning correspondence relationship, and push learning videos and test questions matching the specific needs to the knowledge demander based on the specific needs of the knowledge demander and the knowledge graph.

[0006] Optionally, in S1, based on multiple knowledge points, an optical character recognition method and a large language model are used to process the course video to generate a learning video for each knowledge point, specifically: Based on multiple knowledge points, an optical character recognition method is used to perform text recognition on image information in a course video, and the course video is cut according to the text recognition result to generate a video clip for each knowledge point; The large language model is used to perform text processing on the audio information of each video clip, and text information is added to each video clip based on the text processing results to generate a learning video for each knowledge point.

[0007] Optionally, based on the multiple knowledge points, the optical character recognition method is used to perform text recognition on the image information in the course video, specifically: Decompose the course video into multiple image frames; Based on multiple knowledge points, optical character recognition method is used to perform text recognition on the image information in each image frame.

[0008] Optionally, after generating the video clip of each knowledge point, the method further includes: Based on multiple knowledge points, the course video is segmented using a large language model to generate validation video clips for each knowledge point; Determine the similarity between the video clip of each knowledge point and the verification video clip, and when the similarity is lower than a preset threshold, adjust the cutting position of the corresponding video clip in the course video, and generate the video clip of the corresponding knowledge point again.

[0009] Optionally, the text information includes one or more of video subtitles, video summaries and video tags.

[0010] Optionally, before generating a learning video for each knowledge point, the method further includes: An image frame meeting a preset condition is identified from each video clip using a portrait recognition method, and the image frame is set as a cover of the corresponding video clip.

[0011] Optionally, the large language model is used in S2 to determine the knowledge points in each test question in the question bank, specifically: A large language model is used to analyze each test question in the question bank, and the knowledge points in each test question are determined based on the analysis results.

[0012] Optionally, after constructing the knowledge graph according to the association relationship and the learning relationship, the method further includes: Pushing multiple test questions to the knowledge demander according to the characteristic information of the knowledge demander, and obtaining the answer status of the knowledge demander to the multiple test questions; The specific needs of the knowledge demander are determined according to the answer situation.

[0013] Optionally, determining the association relationship between the multiple knowledge points in S1 is specifically: According to the hierarchical relationship, causal relationship or parallel relationship between every two knowledge points in the multiple knowledge points, the correlation relationship between the multiple knowledge points is determined.

[0014] On the other hand, the present invention also provides a system based on any of the above-mentioned knowledge point learning recommendation methods based on course videos, the system comprising: The video processing module is used to determine the association between multiple knowledge points, and based on the multiple knowledge points, use the optical character recognition method and the large language model to process the course video to generate a learning video for each knowledge point; The learning correspondence module is used to use the large language model to determine the knowledge points in each test question in the question bank, and to determine the learning correspondence between the knowledge points, learning videos, and test questions; The learning push module is used to construct a knowledge graph based on the association relationship and the learning correspondence relationship, and to push learning videos and test questions matching the specific needs to the knowledge demander based on the specific needs of the knowledge demander and the knowledge graph.

[0015] The beneficial effects that the present invention can produce include: The present invention is based on multiple knowledge points, uses optical character recognition methods and large language models to process course videos, and generates learning videos for each knowledge point; then constructs a knowledge graph based on the learning correspondence between knowledge points, learning videos and test questions, as well as the association between multiple knowledge points; finally, based on the knowledge graph and the specific needs of knowledge demanders, pushes learning videos and test questions that match their specific needs to knowledge demanders. This can help knowledge demanders quickly and accurately obtain the required learning content, so that knowledge demanders can learn in a fragmented, precise and flexible manner, thereby meeting the personalized learning needs of knowledge demanders, helping to reduce the learning burden of students, improve students' learning enthusiasm, and achieve the purpose of improving learning efficiency and improving learning effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flowchart of a method for recommending knowledge point learning based on course videos provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is described in detail below in conjunction with embodiments, but the present invention is not limited to these embodiments.

[0018] The embodiment of the present invention provides a knowledge point learning recommendation method based on course videos, such as Figure 1 As shown, the method includes: S1. Determine the correlation between multiple knowledge points, and based on the multiple knowledge points, use the optical character recognition (OCR) method and the large language model to process the course video and generate a learning video for each knowledge point; S2. Use the large language model to determine the knowledge points in each test question in the question bank, and determine the learning correspondence between the knowledge points, learning videos, and test questions; S3. Build a knowledge graph based on association relationships and learning correspondence relationships, and push learning videos and test questions that match the specific needs to knowledge demanders based on their specific needs and knowledge graph.

[0019] Specifically, the learning correspondence means that for each knowledge point, the knowledge point should be associated with its corresponding learning video and corresponding test questions to form a correspondence between knowledge point - learning video - test questions, thereby obtaining segmented learning resources for each knowledge point.

[0020] In this embodiment, the association relationship between multiple knowledge points is determined in S1, which may be: According to the hierarchical relationship, causal relationship or parallel relationship between every two knowledge points in the multiple knowledge points, the correlation relationship between the multiple knowledge points is determined.

[0021] In this embodiment, in S1, based on multiple knowledge points, the course video is processed using the OCR method and the large language model to generate a learning video for each knowledge point, which may be specifically: Based on multiple knowledge points, the OCR method is used to perform text recognition on the image information in the course video, and the course video is cut according to the text recognition results to generate video clips for each knowledge point.

[0022] Furthermore, based on multiple knowledge points, the OCR method is used to perform text recognition on the image information in the course video, which can be specifically: Decompose the course video into multiple image frames; Based on multiple knowledge points, the OCR method is used to perform text recognition on the image information in each image frame.

[0023] Specifically, this embodiment utilizes tools such as FFmpeg to decompose the course video into a series of image frames.

[0024] Specifically, the image information may include PPT text and other text information in the course video. This embodiment uses the OCR method to extract and recognize text information from the image information.

[0025] Specifically, this embodiment cuts the course video according to the text recognition result to generate a video clip for each knowledge point.

[0026] The large language model is used to perform text processing on the audio information of each video clip, and text information is added to each video clip based on the text processing results to generate a learning video for each knowledge point.

[0027] Specifically, the above text information may include one or more of video subtitles, video summaries, and video tags.

[0028] Specifically, text processing includes converting audio information into text and performing in-depth analysis of the text.

[0029] Specifically, this embodiment uses the Automatic Speech Recognition (ASR) method to convert the audio information of each video clip into text, and then uses the Natural Language Processing (NLP) method in the large language model to perform in-depth analysis on these texts, including word segmentation recognition, Named Entity Recognition (NER), syntactic analysis, etc. Then, this embodiment adds video subtitles and video summaries to each video clip based on the in-depth analysis results; at the same time, video tags are added to each video clip, such as chapter name, main concepts covered, applicable test scope, etc., to enhance the searchability of the video clip.

[0030] In this embodiment, before generating a learning video for each knowledge point, the method may further include: The portrait recognition method is used to identify an image frame that meets the preset conditions from each video clip, and the image frame is set as the cover of the corresponding video clip.

[0031] Specifically, the preset conditions may include requirements for the teacher's movements and expressions, etc., such as requiring that the teacher image in the image frame should keep the mouth open, the eyes not closed, and the face clear.

[0032] After the video clips are processed as above, a learning video for each knowledge point can be obtained.

[0033] Furthermore, after generating the video clip of each knowledge point, the method may further include: Based on multiple knowledge points, the course video is segmented using a large language model to generate validation video clips for each knowledge point; Determine the similarity between the video clip of each knowledge point and the validation video clip, and adjust the cutting position of the corresponding video clip in the course video when the similarity is lower than a preset threshold, and generate the video clip of the corresponding knowledge point again.

[0034] Specifically, the large language model can be a mixed-element large model, a listening and understanding large model, etc.

[0035] Specifically, in the course video of this embodiment, the teacher teaches the course in combination with PPT, and each page of the PPT is marked with the knowledge points of that page. This embodiment can use the Hunyuan large model to identify the knowledge points of each page of the PPT, as well as the two image frames corresponding to the start and end time of each page of the PPT, output the start and end time of each page of the PPT, and then cut the course video according to the start and end time of each page of the PPT to obtain the verification video clips for each knowledge point.

[0036] This embodiment uses two methods (a method combining an OCR method with a large language model, and a method using only a large model) to cut the course video, and then compares the cutting effects of the two methods. The cutting position is adjusted according to the comparison results, which can achieve mutual verification of the two methods and effectively avoid the problem of inaccurate cutting caused by a large semantic understanding deviation in the field of traditional Chinese medicine using a single method, so that each video segment focuses on one knowledge point, ensuring the integrity and accuracy of the video segment.

[0037] In this embodiment, the large language model is used in S2 to determine the knowledge points in each test question in the question bank, which may be: A large language model is used to analyze each test question in the question bank, and the knowledge points in each test question are determined based on the analysis results.

[0038] Specifically, the test questions include real exam questions, AI-generated practice questions, etc.

[0039] Specifically, the large language model can perform semantic analysis based on the question stem, options, answers, etc., to identify the knowledge points in the test questions. For example, if the test question involves "classification of mineral Chinese medicines", the large language model can identify that it is an important knowledge point in the classification of mineral Chinese medicines.

[0040] Specifically, after the large language model determines the knowledge points in each test question, this embodiment can verify the knowledge points of each test question to ensure an accurate correspondence between the knowledge points and the test questions.

[0041] Furthermore, this embodiment can determine the learning correspondence between knowledge points, learning videos and test questions based on the learning videos and test questions corresponding to the knowledge points. At the same time, combined with the association between multiple knowledge points, the relationship between knowledge points, learning videos and test questions can be linked to construct a lightweight knowledge graph. The nodes in the knowledge graph represent different knowledge points, and the lines between the nodes represent the relationship between the knowledge points (such as hierarchical relationships, causal relationships, parallel relationships, etc.). In addition, this embodiment can also accurately match the knowledge points with the test points in the syllabus.

[0042] In this embodiment, after constructing the knowledge graph according to the association relationship and the learning relationship, the method may further include: Obtain characteristic information of knowledge demanders; Push multiple test questions to the knowledge demander according to the characteristic information of the knowledge demander, and obtain the knowledge demander's answer status to the multiple test questions; Determine the specific needs of knowledge seekers based on their responses.

[0043] Then, this embodiment pushes learning videos and test questions that match the specific needs to the knowledge demander based on the specific needs and knowledge graph of the knowledge demander.

[0044] Specifically, this embodiment can design a questionnaire on the product side to obtain characteristic information such as the education level, age, whether or not the student has passed the practicing pharmacist examination, whether the job is related to pharmacy, etc. of the knowledge seeker, so as to construct an initial learning profile of the student, including knowledge level, learning goals, time arrangements, etc., so as to tailor differentiated learning plans for the students.

[0045] Specifically, this embodiment can use the Listening and Understanding Model, the Hunyuan Model, etc. to develop a personalized learning plan for students. For example, based on the student's academic qualifications, age, medical examination experience, work background and other characteristic information, for students with a pharmaceutical background, this embodiment can reduce the review of basic knowledge points and directly enter the study of more difficult knowledge points; while for students without a pharmaceutical background, this embodiment can provide more learning videos and test questions on basic knowledge points.

[0046] Specifically, this embodiment can also use the knowledge graph to randomly generate 10-20 knowledge points and push them to students. The large language model can mark the knowledge points that need to be focused on according to the students' answers and feedback. This can not only dynamically adjust the students' learning plans, but also achieve a cold start for new students. For example, if the students' answers on a certain knowledge point are not ideal, this embodiment can increase the push of learning videos and test questions for that knowledge point, and extend the students' learning time for that knowledge point; if the students have already mastered a certain knowledge point, this embodiment can push the learning videos and test questions for that knowledge point, shorten the students' learning time for that knowledge point, and push more learning videos and test questions for other knowledge points that have not been mastered to the students.

[0047] Specifically, this embodiment can also collect students' negative feedback on the learning correspondence between knowledge points, learning videos and test questions, and then adjust the cutting position of the video clips and regenerate the learning video according to the students' negative feedback to adjust the learning correspondence between the three.

[0048] Another embodiment of the present invention provides a system based on any of the above-mentioned knowledge point learning recommendation methods based on course videos, and the system may include: The video processing module is used to determine the association between multiple knowledge points, and based on multiple knowledge points, use the OCR method and large language model to process the course video to generate a learning video for each knowledge point; The learning correspondence module is used to use the large language model to determine the knowledge points in each test question in the question bank, and to determine the learning correspondence between the knowledge points, learning videos, and test questions; The learning push module is used to build a knowledge graph based on association relationships and learning correspondence relationships, and push learning videos and test questions that match the specific needs to knowledge demanders based on their specific needs and knowledge graphs.

[0049] The present invention is based on multiple knowledge points, uses OCR methods and large language models to process course videos, and generates learning videos for each knowledge point; then constructs a knowledge graph based on the learning correspondence between knowledge points, learning videos and test questions, and the association between multiple knowledge points; finally, based on the knowledge graph and the specific needs of knowledge demanders, pushes learning videos and test questions that match their specific needs to knowledge demanders. This can help knowledge demanders quickly and accurately obtain the required learning content, so that knowledge demanders can learn in a fragmented, precise and flexible manner, thereby meeting the personalized learning needs of knowledge demanders, helping to reduce the learning burden of students, improve students' learning enthusiasm, and achieve the purpose of improving learning efficiency and improving learning effects.

[0050] The present invention changes the traditional online course sequential learning and all-learning learning methods, as well as the situation in which traditional online subjects are questions, classes are classes, and test questions and course videos are unrelated, which is not conducive to efficient learning for students. The present invention uses knowledge graphs to conduct closed-loop learning recommendations for students in the form of test-learning-test-mastery, thereby achieving more accurate knowledge point learning recommendations.

[0051] The present invention matches the learning videos of knowledge points with the test questions in the learning path, and can push a small number of test questions to students to evaluate their mastery of multiple knowledge points in the field of medical and nursing learning, thereby providing students with more systematic and accurate learning recommendations, solving the cold start problem of students. The present invention adopts a method combining OCR method and large language model to cut course videos, and also adopts large language model alone to cut course videos, and then verifies the two video cutting results, which can effectively avoid the problem of low accuracy of a single cutting method and improve the matching degree between video clips and knowledge points.

[0052] The above are only a few embodiments of the present application and do not constitute any form of limitation to the present application. Although the present application is disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technician familiar with the profession, without departing from the scope of the technical solution of the present application, using the technical content disclosed above to make slight changes or modifications are equivalent to equivalent implementation cases and fall within the scope of the technical solution.

Claims

1. A knowledge point learning recommendation method based on course videos, characterized in that: The method comprises: S1. Determine the correlation between multiple knowledge points, and based on the multiple knowledge points, use the optical character recognition method and the large language model to process the course video to generate a learning video for each knowledge point; S2. Use the large language model to determine the knowledge points in each test question in the question bank, and determine the learning correspondence between the knowledge points, learning videos, and test questions; S3. Construct a knowledge graph based on the association relationship and the learning correspondence relationship, and push learning videos and test questions matching the specific needs to the knowledge demander based on the specific needs of the knowledge demander and the knowledge graph.

2. The method according to claim 1, characterized in that In S1, based on multiple knowledge points, the course video is processed using an optical character recognition method and a large language model to generate a learning video for each knowledge point, specifically: Based on multiple knowledge points, an optical character recognition method is used to perform text recognition on image information in a course video, and the course video is cut according to the text recognition result to generate a video clip for each knowledge point; The large language model is used to perform text processing on the audio information of each video clip, and text information is added to each video clip based on the text processing results to generate a learning video for each knowledge point.

3. The method according to claim 2, characterized in that Based on multiple knowledge points, the optical character recognition method is used to perform text recognition on the image information in the course video, specifically: Decompose the course video into multiple image frames; Based on multiple knowledge points, optical character recognition method is used to perform text recognition on the image information in each image frame.

4. The method according to claim 2, characterized in that: After generating the video clip of each knowledge point, the method further includes: Based on multiple knowledge points, the course video is segmented using a large language model to generate validation video clips for each knowledge point; Determine the similarity between the video clip of each knowledge point and the verification video clip, and when the similarity is lower than a preset threshold, adjust the cutting position of the corresponding video clip in the course video, and generate the video clip of the corresponding knowledge point again.

5. The method according to claim 2, characterized in that: The text information includes one or more of video subtitles, video summaries, and video tags.

6. The method according to claim 2, characterized in that Before generating the learning video for each knowledge point, the method further includes: An image frame meeting a preset condition is identified from each video clip using a portrait recognition method, and the image frame is set as a cover of the corresponding video clip.

7. The method according to claim 1, characterized in that In S2, the large language model is used to determine the knowledge points in each test question in the question bank, specifically: Use a large language model to analyze each test question in the question bank, and determine the knowledge points in each test question based on the analysis results.

8. The method according to claim 1, characterized in that After constructing the knowledge graph according to the association relationship and the learning relationship, the method further includes: Pushing multiple test questions to the knowledge demander according to the characteristic information of the knowledge demander, and obtaining the answer status of the knowledge demander to the multiple test questions; The specific needs of the knowledge demander are determined according to the answer situation.

9. The method according to claim 1, characterized in that: The association relationship between multiple knowledge points is determined in S1, specifically: According to the hierarchical relationship, causal relationship or parallel relationship between every two knowledge points in the multiple knowledge points, the correlation relationship between the multiple knowledge points is determined.

10. A system based on the knowledge point learning recommendation method based on course videos according to any one of claims 1 to 9, characterized in that: The system comprises: The video processing module is used to determine the association between multiple knowledge points, and based on the multiple knowledge points, use the optical character recognition method and the large language model to process the course video to generate a learning video for each knowledge point; The learning correspondence module is used to use the large language model to determine the knowledge points in each test question in the question bank, and to determine the learning correspondence between the knowledge points, learning videos, and test questions; The learning push module is used to construct a knowledge graph based on the association relationship and the learning correspondence relationship, and to push learning videos and test questions matching the specific needs to the knowledge demander based on the specific needs of the knowledge demander and the knowledge graph.

Images