Teaching method and system based on digital media

Abstract

The invention discloses a teaching method and system based on digital media, relates to the technical field of digital media teaching, and solves the technical problem that the learning effect of a user is poor due to the fact that personalized teaching service is not performed on the user in the prior art. Education resources of different education levels and identity information of users are collected, the education resources are graded based on the education levels to obtain level education resources, and the identity information is matched with the level education resources to obtain corresponding user level education resources; according to the error induction in the identity information, matching with the matching following resource to obtain an error type in the matching following resource, matching the error type with an error type table to obtain an error-prone type, and combining with the error resource to obtain a recommended educational resource; and after the user completes learning of the recommended education resources, collecting a feedback result of the user, calculating a resource score according to the feedback result, and optimizing the recommended education resources based on the resource score.

Description

Technical Field

[0001] The invention belongs to the field of digital media teaching, and specifically relates to a teaching method and system based on digital media. Background Art

[0002] Digital media refers to the medium form of transmitting information through computer networks, mobile devices and other digital technology means. It covers a variety of types such as text, images, audio, video and interactive content; in the field of education, the application of digital media has greatly enriched teaching methods and learning experience, and provided strong support for personalized learning.

[0003] The invention patent with application number CN2012105097288 discloses a multimedia digital teaching resource system and its establishment method. The invention pre-processes the collected teaching data and stores it in a relational database. The data stored in the relational database is classified into standardized courses and subjects, the quality of the collection is quantitatively evaluated, and the knowledge points are cross-linked. Users can search and analyze the relational database and perform learning interactions and other operations. When classifying teaching resources, the invention classifies them according to subjects and courses. For individual users, they can only find teaching resources for the corresponding courses by themselves, and cannot conduct effective targeted learning for users, which will result in users being unable to consolidate and break through their weaknesses during learning, resulting in poor learning effects.

[0004] The present invention provides a teaching method and system based on digital media to solve the above technical problems. Summary of the invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes a teaching method and system based on digital media, which are used to solve the technical problem in the prior art that no personalized teaching service is provided to users, resulting in poor learning effect for users.

[0006] To achieve the above object, a first aspect of the present invention provides a teaching method based on digital media, comprising:

[0007] Step 1: Collect and store educational resources of different educational levels and user identity information;

[0008] Step 2: Based on the education level, the educational resources are classified to obtain the graded educational resources; according to the identity information, the graded educational resources are matched to complete the progress tracking to obtain the matched follow-up resources;

[0009] Step 3: Recommend educational resources for users based on matching follow-up resources;

[0010] Step 4: Conduct quality assessment on recommended educational resources to obtain resource scores; optimize recommended educational resources based on the resource scores.

[0011] It should be noted that the collected educational resources are all teaching materials that have been quality reviewed by subject experts.

[0012] Preferably, the step of grading educational resources based on educational levels to obtain graded educational resources includes:

[0013] A1: Extracting educational levels and educational resources; educational resources include teaching courses and teaching materials extracted from online course platforms; educational levels include grades and subjects;

[0014] A2: Match the grades in the education level with the subjects and educational resources, and determine whether the teaching courses and teaching materials in the educational resources correspond to the grades and subjects; if yes, add the corresponding grades and subjects as grade and subject labels to the educational resources to obtain the grade and subject stage educational resources; if no, jump to A3;

[0015] A3: Divide educational resources according to the subjects they belong to, and combine the corresponding subjects as level labels with educational resources to obtain level-expanded educational resources;

[0016] A4: Mark the subject-level educational resources and subject-level extended educational resources as subject-level educational resources.

[0017] Preferably, the matching of identity information with the subject education resources to complete progress following and obtain matching following resources includes:

[0018] Extract the user's identity information; the identity information includes the user's learning level, historical learning records, and wrong question summary;

[0019] Match the learning level with the level and subject label of the level and subject education resource to obtain the corresponding user level and subject education resource; extract the learning time corresponding to the historical learning record, and select the historical learning record with the closest learning time to the current time as the progress record;

[0020] The progress records are refined to obtain progress knowledge points, the user-level science education resources are traversed and the number of occurrences of progress knowledge points in each chapter is counted, and the chapters with the most occurrences of progress knowledge points are selected and marked as progress chapters; the education resources corresponding to the progress chapters are marked as matching follow-up resources.

[0021] Preferably, the step of extracting the progress records to obtain progress knowledge points includes:

[0022] Establish corresponding subject knowledge graphs for different grades and subjects;

[0023] The keywords in the progress record are extracted through natural language processing technology, and the keywords with the same semantics are combined to obtain synonymous keywords; it is determined whether the synonymous keywords match the nodes in the subject knowledge graph; if so, the knowledge point of the corresponding node is marked as a progress knowledge point; if not, the corresponding synonymous keyword is marked as a regular keyword.

[0024] Preferably, the step of recommending recommended educational resources to the user based on matching follow-up resources includes:

[0025] Extract the wrong questions from the matching follow-up resources and user identity information;

[0026] Extract wrong question knowledge points and wrong question types in wrong question summary through natural language processing technology, match wrong question knowledge points with matching follow-up resources to obtain wrong question resources;

[0027] Define a wrong question type table, and compare the wrong question types with the wrong question type table one by one, count the number of occurrences of each wrong question type, sort the wrong question types from large to small according to the number of occurrences, and select the top n wrong question types to mark as the user's easy-to-make mistakes;

[0028] Recommended educational resources are obtained by combining easy-to-make mistakes types, wrong question resources and matching follow-up resources.

[0029] Preferably, the step of combining the error-prone types, the error question resources and the matching follow-up resources to obtain the recommended educational resources includes:

[0030] Extract error-prone types, wrong question resources, and matching follow-up resources;

[0031] Divide the matching and following resources into several groups of matching type resources according to the question type; mark the resources with the same question type and easy-to-make mistakes type in the wrong question resources as consolidation resources;

[0032] Set the type coefficient according to the error-prone type; by the formula LXSi=LYXi×CTSi×[CY+(1-CY)e -CTi / CS ] Calculate the priority coefficient LXSi of the consolidation resource numbered i; where LYXi represents the type coefficient of the error-prone type corresponding to the consolidation resource numbered i, CTSi represents the number of occurrences of the corresponding question of the error-prone type, CY represents the retention ratio of memory retention, and 0<CY<1, CTi represents the time from the last occurrence of the question corresponding to the consolidation resource numbered i to the current time, and CS represents the speed parameter of memory decay; i=1, 2, 3……n, n is a positive integer; the retention ratio and speed parameter are evaluated by experts;

[0033] Sort the consolidation resources according to their priority coefficients, compare the question types corresponding to the n highest ranked consolidation resources with the error-prone types, and determine whether the question types corresponding to the consolidation resources are the same as the error-prone types; if yes, use the corresponding matching type resources as recommended educational resources; if not, use the corresponding matching type resources as normal progress resources.

[0034] Preferably, setting the type coefficient according to the error-prone type includes:

[0035] Extract the error-prone types and the corresponding average score ratios from the online course platform, and count the correct answer rates and answer time of the error-prone types for platform users;

[0036] Set the score weight coefficient for the average score ratio, set the accuracy weight coefficient for the answer accuracy, and set the time weight coefficient for the answer time; calculate the type coefficient by calculating the product of the average score ratio and the score weight coefficient, the product of the answer accuracy and the accuracy weight coefficient, and 1 minus the product of the answer time and the time weight coefficient; among which, the score weight coefficient, the accuracy weight coefficient, and the time weight coefficient add up to one, and are set according to the importance of the average score ratio, the answer accuracy, and the answer time to the mastery of the knowledge points.

[0037] Preferably, the quality assessment of the recommended educational resources to obtain a resource score includes:

[0038] Collect user feedback results; the feedback results include feedback accuracy and feedback time;

[0039] The resource score is calculated by calculating the ratio of the product of the feedback accuracy and the accuracy weight coefficient to the product of the feedback time and the time weight coefficient.

[0040] Preferably, the optimizing the recommended educational resources based on resource ratings includes:

[0041] Extract resource scores; set recommendation optimization thresholds; where the recommendation optimization thresholds are set based on the experts’ required mastery of the corresponding subject for the corresponding grade;

[0042] Determine whether the resource score is greater than the optimization threshold; if yes, mark the optimization result as not requiring optimization; if no, mark the optimization result as requiring optimization, annotate the knowledge points and question types corresponding to the recommended educational resources, and add them to the wrong question resources.

[0043] A digital media-based teaching system, applied to the above-mentioned digital media-based teaching method, characterized by comprising: a data collection module, a resource matching module, a display module and a resource evaluation module;

[0044] Data collection module: used to collect and store educational resources of different education levels and user identity information

[0045] Resource matching module: used to classify educational resources based on educational levels to obtain graded educational resources; match the graded educational resources according to identity information to complete progress tracking to obtain matching follow-up resources; recommend recommended educational resources to users based on matching follow-up resources;

[0046] Display module: displays the recommended educational resources on the user interface;

[0047] Resource evaluation module: used to evaluate the quality of recommended educational resources to obtain resource scores; and optimize recommended educational resources based on resource scores.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. The present invention collects educational resources of different educational levels and the identity information of users, classifies educational resources based on educational levels to obtain subject-level educational resources, matches identity information with subject-level educational resources to obtain corresponding user-level subject-level educational resources, selects progress records, and determines matching follow-up resources by counting the number of occurrences of progress knowledge points in the progress records, thereby ensuring that the pushed educational resources are highly relevant, accurately matching the required educational resources for users, and improving the personalized experience of users; matching the wrong question summary in the identity information with the matching follow-up resources to obtain the wrong question type in the matching follow-up resources, matching the wrong question type with the wrong question type table to obtain the easy-to-make type, and combining the wrong question resources to obtain recommended educational resources, which can provide more personalized tutoring, help students quickly overcome learning obstacles, and improve the effectiveness and pertinence of learning. It can also discover common learning obstacles through analysis of a large amount of user data, and help improve teaching content and methods; after the user completes the study of the recommended educational resources, the user's feedback results are collected, the resource score is calculated according to the feedback results, and the recommended educational resources are optimized based on the resource score, so that the user's learning effect is more perfect.

[0050] 2. The present invention divides the matching follow-up resources according to the question type to obtain matching type resources, and marks the resources with the same question type and easy-to-make mistakes type in the matching type resources as consolidation resources; sets the corresponding type coefficient according to the easy-to-make mistakes type, and different types of questions have different effects on improving the user's learning effect. In combination with the memory retention ratio and the memory decay speed parameters, the priority coefficient of the consolidation resource is calculated, and the consolidation resources are sorted according to the priority coefficient to obtain recommended educational resources in which the user is more likely to make mistakes. The wrong questions and easy-to-make mistakes types are accurately analyzed, which improves the effectiveness and pertinence of learning. According to the feedback from the system, the weak links of students can be understood, and more pertinent guidance and support can be provided to help improve learning methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 The following is a flowchart of an embodiment of the present invention.

[0053] Figure 2 It is a system composition diagram of the present invention.

[0054] Figure 3 The complete flow chart of obtaining subject education resources in one embodiment of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] See also Figure 1-Figure 3 The first aspect of the present invention provides a digital media-based teaching method, comprising:

[0057] Step 1: Collect and store educational resources of different education levels and user identity information.

[0058] For example, teaching courses and teaching materials for different grades and subjects are collected from online course platforms, such as primary and secondary school course videos, CET-4 and CET-6 English materials, or professional skills training lectures, and other educational resources for different learning stages. The educational resources are learning materials and corresponding test questions for explaining or learning certain knowledge points. The user's learning stage, learning progress, and a summary of wrong questions on the teaching platform are collected through the user's platform registration information.

[0059] Step 2: Based on the education level, the educational resources are classified to obtain the graded educational resources; according to the identity information, the graded educational resources are matched to complete the progress following to obtain the matched follow-up resources.

[0060] Exemplarily, the collected educational resources are divided according to different learning stages to determine whether the educational resources correspond to grades and subjects, that is, whether they belong to the primary and secondary school stage or the university stage. Both the primary and secondary school stage and the university stage have corresponding grades and subjects, while extended education does not include grades, only corresponding subjects. The grades and subjects of the primary and secondary school stage and the university stage with corresponding grades and subjects are added as grade and subject labels to the corresponding educational resources to obtain grade and subject stage educational resources. The educational resources corresponding to the extended education are divided according to their subjects, and the corresponding subjects are combined with the corresponding educational resources as grade and subject labels to obtain grade and subject extended educational resources; the grade and subject extended educational resources are combined with the grade and subject stage educational resources to obtain grade and subject educational resources. The system can more accurately provide personalized resource recommendations based on the user's grade and subject, thereby improving the versatility and applicability of the system.

[0061] The learning level in the user's identity information is matched with the level label of the level education resource. In this embodiment, the user's learning level is the second year of high school, and the corresponding education resources are matched. The user's historical learning records are extracted from the online course platform, and the corresponding learning time is extracted. The historical learning records with the closest learning time to the current time are selected as progress records; in this embodiment, the historical learning records closest to the current time are the learning "sets", indicating that the user has recently been learning related content of the "set".

[0062] In some other preferred embodiments, if the user's learning level belongs to the subject expansion educational resources, the corresponding educational resources are matched according to the user's subject. If the user's subject is a legal subject, the user is matched with educational resources related to the legal examination.

[0063] Corresponding subject knowledge graphs are established for different grades and different subjects. The subject knowledge graph contains a summary of all knowledge points of the corresponding grade and subject. Different nodes in the subject knowledge graph correspond to different knowledge points. The corresponding subject knowledge graph can be obtained according to the grade and subject matching; the keywords in the progress record are refined through natural language processing technology. The keywords are words related to the learning content in the progress record, and all keywords with the same semantics are combined to obtain synonymous keywords; it is determined whether the synonymous keywords match the nodes in the subject knowledge graph, and the synonymous keywords that can match the nodes are marked as progress knowledge points, indicating that the synonymous keywords are the corresponding subject knowledge points. Through in-depth analysis of the progress records, the system can more accurately identify the user's current learning status and needs, and provide highly relevant learning resources.

[0064] In the progress record of learning "collection", several knowledge points related to collection are extracted, and the user-level education resources are traversed to count the number of times different knowledge points appear in each chapter. It is found that the knowledge point appears 25 times in the content of Chapter 3, 8 times in the content of Chapter 4, and 7 times in the content of Chapter 5. The content of Chapter 3 is an education resource that mainly talks about the content of "collection", and the content of Chapter 4 and Chapter 5 is also related to "collection". Therefore, Chapter 3 is the chapter that the user is studying recently. Mark Chapter 3 as the user's progress chapter, and mark the education resources corresponding to the content of Chapter 3 as matching follow-up resources to ensure that the provided resources meet the user's current learning needs, improve the effectiveness and satisfaction of learning, and more accurately match the required resource needs for users.

[0065] In some other preferred embodiments, the user may have completed the historical learning record with the learning time closest to the current time, and has not yet started the next stage of learning. It is necessary to analyze the user's specific progress in the progress record more carefully; further refine the progress knowledge points, and extract the progress knowledge points closest to the current time according to the learning time, match the progress knowledge points in the progress chapter, and obtain the ranking completion of the progress knowledge points in the progress chapter. The ranking completion is the proportion of the position of the progress knowledge points in the progress chapter. According to the knowledge point progress, it is judged whether the user has completed learning the content of this chapter. If the learning of the chapter content has been completed, the educational resources corresponding to the next chapter will be used as matching follow-up resources; if the learning of the chapter content has not been completed, the educational resources corresponding to the corresponding chapter will be used as matching follow-up resources. This can more accurately identify the user's current learning status and needs, provide highly relevant learning resources, and adjust the learning plan at any time according to one's personal needs.

[0066] Step 3: Recommend educational resources for users based on matching follow-up resources.

[0067] Exemplarily, natural language processing technology is used to extract wrong question knowledge points and wrong question types corresponding to wrong questions in the user's wrong question summary, and the wrong question knowledge points are matched with matching follow-up resources to obtain corresponding wrong question resources, which are learning materials corresponding to the user's wrong question knowledge points; this embodiment defines a wrong question type table, which contains statistics of all question types of the user in the high school mathematics subject, including multiple-choice questions, fill-in-the-blank questions, graphic analysis questions, and formula calculation questions; the user's wrong question type is matched with the wrong question type table, and the number of occurrences of each type in the wrong question type table is counted, and sorted from large to small according to the number of occurrences, and the top two wrong question types are selected as the user's easy-to-make mistakes types. In this embodiment, the user's easy-to-make mistakes types are fill-in-the-blank questions and graphic analysis questions.

[0068] The easy-to-make-mistake types and the corresponding average score ratios are extracted through the online course platform, and the average score ratio is the average ratio of the corresponding question types to the test score; the correct answer rate and answer time of users on similar questions in the online course platform records are counted, and the answer time is standardized to obtain the standard answer time; in this embodiment, the score weight coefficient set for the average score ratio, the correct answer rate and the standard answer time is 0.5, the correct rate weight coefficient is 0.3, and the time weight coefficient is 0.2; the type coefficient LYX is calculated by the formula LYX=FQS×FSB+ZQS×DTL+tTQS×(1-HDT); wherein FQS represents the score weight coefficient, FSB represents the average score ratio, 0<FSB<1, ZQS represents the correct rate weight coefficient, DTL represents the correct answer rate, TQS represents the time weight coefficient, HDT represents the standard answer time, and FQS+ZQS+TQS=1.

[0069] The consolidation resources are numbered as i, i = 1, 2, 3...n, and the formula LXSi = LYXi × CTSi × [CY + (1-CY)e -CTi / CS ] Calculate the priority coefficient LXSi of the consolidation resource numbered i; sort the consolidation resources according to the priority coefficients from large to small, compare the question types corresponding to the first two consolidation resources with the error-prone types, and determine whether the question types corresponding to the consolidation resources are the same as the error-prone types. In this embodiment, it is determined that the question type corresponding to the highest-ranked consolidation resource is a graphic analysis question, which is the same as the error-prone type. Then, the corresponding matching type resource is used as a recommended educational resource, a recommendation window is created, and the recommended educational resources are displayed in the recommendation window, providing more personalized tutoring, accurately analyzing wrong questions and error-prone types, improving the effectiveness and pertinence of learning, understanding the user's weak links, providing more targeted tutoring and support, and helping to improve learning methods.

[0070] It should be noted that the recommended educational resources are the knowledge points that the user has made mistakes in the past learning process, and the user's memory of the knowledge points is calculated based on the elapsed time, memory retention ratio and memory decay. The longer the time, the more vague the user's memory of the knowledge points, but the knowledge points after learning will be retained in the memory, and timely recall and review will help users consolidate the knowledge points.

[0071] The recommended educational resources and matching follow-up resources are sent to the user end, and the user learns the matching follow-up resources and provides feedback and evaluation on the recommended educational resources.

[0072] Step 4: Conduct quality assessment on recommended educational resources to obtain resource scores; optimize recommended educational resources based on the resource scores.

[0073] Exemplarily, user feedback results on recommended educational resources are collected, and the feedback results include the user's feedback accuracy and feedback time on test questions in the recommended educational resources, and the ratio of the product of the feedback accuracy rate and the accuracy weight coefficient to the product of the feedback time and the time weight coefficient is used as the resource score; in this embodiment, the recommendation optimization threshold is set to 1.5 to determine whether the resource score is greater than the optimization threshold; in this embodiment, the resource score is 2, which is greater than the optimization threshold, indicating that the user's consolidation effect on the recommended educational resources has achieved the expected effect, and there is no need to optimize the recommended educational resources.

[0074] In some other preferred embodiments, if the resource score is less than the optimization threshold, it means that after the user has consolidated the recommended educational resources, the effect still does not meet the expectations. In this case, the knowledge points and question types corresponding to the recommended educational resources will be marked and added to the next wrong question resources.

[0075] A digital media-based teaching system, used for the digital media-based teaching method, comprising: a data collection module, a resource matching module, a display module and a resource evaluation module;

[0076] Data collection module: used to collect and store educational resources of different education levels and user identity information

[0077] Resource matching module: used to classify educational resources based on educational levels to obtain graded educational resources; match the graded educational resources according to identity information to complete progress tracking to obtain matching follow-up resources; recommend recommended educational resources to users based on matching follow-up resources;

[0078] Display module: displays the recommended educational resources on the user interface;

[0079] Resource evaluation module: used to evaluate the quality of recommended educational resources to obtain resource scores; and optimize recommended educational resources based on resource scores.

[0080] Part of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is a formula closest to the actual situation obtained by software simulation of a large amount of collected data; the preset parameters and preset thresholds in the formula are set by technical personnel in this field according to actual conditions or obtained through simulation of a large amount of data.

[0081] Working principle of the present invention:

[0082] The present invention collects educational resources of different educational levels and identity information of users, classifies educational resources based on educational levels to obtain subject-level educational resources, matches identity information with subject-level educational resources to obtain corresponding user-level subject educational resources, and determines matching follow-up resources by counting the number of occurrences of progress knowledge points in progress records; matches the wrong question summary in the identity information with the matching follow-up resources to obtain the wrong question type in the matching follow-up resources, matches the wrong question type with the wrong question type table, divides the matching follow-up resources according to the question type to obtain matching type resources, and marks the resources with the same question type and easy-to-make mistakes type in the matching type resources as consolidation resources; sets the corresponding type coefficient according to the easy-to-make mistakes type, calculates the priority coefficient of the consolidation resources, and sorts the consolidation resources according to the priority coefficient to obtain recommended educational resources in which the user is more likely to make mistakes; collects user feedback results, calculates resource scores according to the feedback results, and optimizes the recommended educational resources based on the resource scores.

[0083] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A teaching method based on digital media, characterized in that: include: Step 1: Collect and store educational resources of different educational levels and user identity information; Step 2: Based on the education level, the educational resources are classified to obtain the graded educational resources; according to the identity information, the graded educational resources are matched to complete the progress tracking to obtain the matched follow-up resources; Step 3: Recommend educational resources for users based on matching follow-up resources; Step 4: Conduct quality assessment on recommended educational resources to obtain resource scores; optimize recommended educational resources based on the resource scores.

2. A digital media-based teaching method according to claim 1, characterized in that: The step of grading educational resources based on educational levels to obtain graded educational resources includes: A1: Extracting educational levels and educational resources; educational resources include teaching courses and teaching materials extracted from online course platforms; educational levels include grades and subjects; A2: Match the grades in the education level with the subjects and educational resources, and determine whether the teaching courses and teaching materials in the educational resources correspond to the grades and subjects; if yes, add the corresponding grades and subjects as grade and subject labels to the educational resources to obtain the grade and subject stage educational resources; if no, jump to A3; A3: Divide educational resources according to the subjects they belong to, and combine the corresponding subjects as level labels with educational resources to obtain level-expanded educational resources; A4: Mark the subject-level educational resources and subject-level extended educational resources as subject-level educational resources.

3. A digital media-based teaching method according to claim 1, characterized in that: The matching of identity information with the subject education resources to complete the progress following and obtain the matching following resources includes: Extract the user's identity information; the identity information includes the user's learning level, historical learning records, and wrong question summary; Match the learning level with the level label of the level education resource to obtain the corresponding user level education resource; extract the learning time corresponding to the historical learning record, and select the historical learning record with the closest learning time to the current time as the progress record; The progress records are refined to obtain progress knowledge points, the user-level science education resources are traversed and the number of occurrences of progress knowledge points in each chapter is counted, and the chapters with the most occurrences of progress knowledge points are selected and marked as progress chapters; the education resources corresponding to the progress chapters are marked as matching follow-up resources.

4. A digital media-based teaching method according to claim 3, characterized in that: The process of refining the progress records to obtain progress knowledge points includes: Establish corresponding subject knowledge graphs for different grades and subjects; The keywords in the progress record are extracted through natural language processing technology, and the keywords with the same semantics are combined to obtain synonymous keywords; it is determined whether the synonymous keywords match the nodes in the subject knowledge graph; if so, the knowledge point of the corresponding node is marked as a progress knowledge point; if not, the corresponding synonymous keyword is marked as a regular keyword.

5. The digital media-based teaching method according to claim 1, characterized in that: The method of recommending recommended educational resources to the user based on matching and following resources includes: Extract the wrong questions from the matching follow-up resources and user identity information; Extract wrong question knowledge points and wrong question types in wrong question summary through natural language processing technology, match wrong question knowledge points with matching follow-up resources to obtain wrong question resources; Define a wrong question type table, and compare the wrong question types with the wrong question type table one by one, count the number of occurrences of each wrong question type, sort the wrong question types from large to small according to the number of occurrences, and select the top n wrong question types to mark as the user's easy-to-make mistakes; Recommended educational resources are obtained by combining easy-to-make mistakes types, wrong question resources and matching follow-up resources.

6. A digital media-based teaching method according to claim 5, characterized in that: The recommended educational resources obtained by combining the error-prone types, wrong question resources and matching follow-up resources include: Extract error-prone types, wrong question resources, and matching follow-up resources; Divide the matching and following resources into several groups of matching type resources according to the question type; mark the resources with the same question type and error-prone type in the matching type resources as consolidation resources; Set the type coefficient according to the error-prone type; by the formula LXSi=LYXi×CTSi×[CY+(1-CY)e -CTi / CS ] Calculate the priority coefficient LXSi of the consolidation resource numbered i; where LYXi represents the type coefficient of the error-prone type corresponding to the consolidation resource numbered i, CTSi represents the number of occurrences of the corresponding question of the error-prone type, CY represents the retention ratio of memory retention, and 0<CY<1, CTi represents the time from the last occurrence of the question corresponding to the consolidation resource numbered i to the current time, and CS represents the speed parameter of memory decay; i=1, 2, 3……n, n is a positive integer; the retention ratio and speed parameter are evaluated by experts; Sort the consolidation resources according to their priority coefficients, compare the question types corresponding to the n highest ranked consolidation resources with the error-prone types, and determine whether the question types corresponding to the consolidation resources are the same as the error-prone types; if yes, use the corresponding matching type resources as recommended educational resources; if not, use the corresponding matching type resources as normal progress resources.

7. A digital media-based teaching method according to claim 6, characterized in that: The step of setting the type coefficient according to the error-prone type includes: Extract the error-prone types and the corresponding average score ratios from the online course platform, and count the correct answer rates and answer time of the error-prone types for platform users; Set a score weight coefficient for the average score ratio, set a correctness weight coefficient for the answer accuracy, and set a time weight coefficient for the answer time; calculate the type coefficient by calculating the product of the average score ratio and the score weight coefficient, the product of the answer accuracy and the correctness weight coefficient, and 1 minus the product of the answer time and the time weight coefficient; among which, the sum of the score weight coefficient, the correctness weight coefficient, and the time weight coefficient is equal to one.

8. The digital media-based teaching method according to claim 1, characterized in that: The quality assessment of the recommended educational resources to obtain a resource score includes: Collect user feedback results; the feedback results include feedback accuracy and feedback time; The resource score is calculated by calculating the ratio of the product of the feedback accuracy and the accuracy weight coefficient to the product of the feedback time and the time weight coefficient.

9. The digital media-based teaching method according to claim 1, characterized in that: The optimization of recommended educational resources based on resource ratings includes: Extract resource scores; set recommendation optimization thresholds; where the recommendation optimization thresholds are set based on the experts’ required mastery of the corresponding subject for the corresponding grade; Determine whether the resource score is greater than the optimization threshold; if yes, mark the optimization result as not requiring optimization; if no, mark the optimization result as requiring optimization, annotate the knowledge points and question types corresponding to the recommended educational resources, and add them to the wrong question resources.

10. A digital media-based teaching system, used to run a digital media-based teaching method according to any one of claims 1 to 9, characterized in that: include: Data collection module, resource matching module, display module and resource evaluation module; Data collection module: used to collect and store educational resources of different education levels and user identity information Resource matching module: used to classify educational resources based on educational levels to obtain graded educational resources; match the graded educational resources according to identity information to complete progress tracking to obtain matching follow-up resources; recommend recommended educational resources to users based on matching follow-up resources; Display module: displays the recommended educational resources on the user interface; Resource evaluation module: used to evaluate the quality of recommended educational resources to obtain resource scores; and optimize recommended educational resources based on resource scores.

Images