Method for automatically generating knowledge point electronic teaching plan based on knowledge graph

Abstract

The invention relates to the technical field of interactive network teaching, in particular to a method for automatically generating a knowledge point electronic teaching plan based on a knowledge graph. A method for automatically generating a knowledge point electronic teaching plan based on a knowledge graph comprises the following steps: an extraction step, a matching step, a training step, a teaching plan generation step, and a final generation of a total teaching plan text with branch teaching plan contents and branch teaching paths. According to the technical scheme, a demand command is generated according to a teaching plan demand input by a teacher, targeted content extraction is carried out, keywords are extracted from three dimensions of a text, namely a theoretical dimension, a material dimension and a template dimension, attribute classification is carried out, available data capable of being carried out by big data are formed, relation logic is extracted, and a knowledge point graph is generated. After learning is carried out through the big data model, the teaching plan is further optimized to obtain a total teaching plan conforming to the teaching process so as to cope with different temporary requirements in a classroom, and coping teaching paths and case contents can be provided after various temporary conditions are encountered.

Description

Technical Field

[0001] The present invention relates to the technical field of interactive network teaching, and in particular to a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph. Background Art

[0002] The rapid development of artificial intelligence big model technology has attracted widespread attention around the world. With powerful natural language processing capabilities, big models can complete complex tasks such as problem solving, content creation, and code generation, showing great potential, which may have a profound impact on human information acquisition methods, knowledge structure, and education models. However, these general big models are not good at solving professional education problems. From general big models to special big models in the field of education, it is an inevitable trend for the deepening development of artificial intelligence big model technology. The education big model is not a fine-tuning and optimization based on the general big model, but a systematic change with the goal of reconstructing the future education landscape, based on an open algorithm model architecture, and centered on innovative education application scenarios. How to clarify the concept and connotation of the education big model and design the system architecture based on the essence of technology to further create new scenarios for education applications has become an issue of the times related to the digital transformation and intelligent upgrading of education.

[0003] Today, AI big models have been widely accepted and supported by the education community. However, at the current stage, few individual teachers can use AI big models to reduce the burden of teaching and improve efficiency. Although AI lesson preparation provides many advantages, it also has some disadvantages and challenges. AI lesson preparation is highly dependent on technology. Once a technical failure or algorithm error occurs, it may affect the teaching quality and lesson preparation efficiency. At the same time, the AI ​​lesson preparation system may have certain limitations and it is difficult to cope with all teaching scenarios and needs, especially when innovation and flexibility are required. Summary of the invention

[0004] In view of the deficiencies in the above-mentioned prior art, the object of the present invention is to provide a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph, which protects the electronic components arranged on the rear end cover of the generator while taking into account the good heat dissipation required by the generator.

[0005] To achieve the above object, the present invention provides the following technical solutions: A method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph comprises the following steps: An extraction step, receiving a demand command including teaching demand content and identity information, extracting theoretical keywords and material keywords in the demand content text, associating the theoretical keywords with the material keywords to form a relationship set, and obtaining a knowledge relationship map; A matching step, filling the knowledge relationship map into a matching teaching plan template to form at least one teaching plan outline, searching for extended content matching the material keyword, and adding the extended content into the teaching plan outline to form a teaching plan text; A training step, generating usage attribute parameters based on the identity information, wherein the usage attribute parameters at least include a teaching object parameter, a teaching purpose parameter, and a teaching scenario parameter, and inputting the teaching plan text into a large model for training and classification to obtain a teaching plan set with different usage attribute classifications; The lesson plan generation step is to merge the content of the lesson plan texts in the lesson plan set to form a simplified lesson plan text, unify and integrate multiple simplified lesson plan texts, form branch lesson plan contents at different points of teaching logic relationships, and form a total lesson plan text with the branch lesson plan contents and branch teaching paths.

[0006] As a preferred embodiment of the technical solution of the present invention, the demand command includes identity attribute information and the demand content text, the identity attribute information includes identity ID, identity information, teaching schedule information, teaching object information, teaching purpose information and teaching scenario information; the demand content text includes theoretical content, material content and template content.

[0007] As a preferred embodiment of the technical solution of the present invention, the extraction step specifically includes: A receiving sub-step, receiving the demand command including teaching demand content and identity information; A classification sub-step, extracting the demand content text, classifying the demand content text by attributes according to different attribute tags, and obtaining a theoretical content text, a material content text and a template text; The keyword extraction sub-step is to identify the theoretical content text, the material content text and the template text information, and obtain the theoretical keywords, the material keywords and the template number respectively; The graph construction sub-step associates the theoretical keywords with the material keywords, extracts the keyword entities and knowledge point relationships between the theoretical keywords and the material keywords, and generates the knowledge relationship graph.

[0008] As a preferred embodiment of the technical solution of the present invention, the feature vector of the material keyword is analyzed based on the theoretical keyword, the matching material supplementary material is queried according to the feature vector, the material supplementary material is filled into the material keyword, and a branch theoretical keyword of the theoretical keyword is generated, and the branch theoretical keyword is associated with the material keyword to update the knowledge relationship map.

[0009] As a preferred embodiment of the technical solution of the present invention, the matching step specifically includes: A matching sub-step, retrieving a corresponding teaching plan template from a teaching plan template library according to the template number; A generation sub-step, filling the knowledge relationship map into the teaching plan template to form the teaching plan outline; The expansion sub-step is to search for expansion contents matching the material keywords, and add the expansion contents into the teaching plan outline to form the teaching plan text.

[0010] As a preferred embodiment of the technical solution of the present invention, all the lesson plan template entities are identified to obtain the module entities in each module text in the template, the process logic relationship between the module entities is extracted according to the context relationship, and the module entities are compared with the keyword entities for similarity to obtain the most suitable lesson plan template.

[0011] As a preferred embodiment of the technical solution of the present invention, in the training step: The usage attribute parameters are generated according to the identity information, the lesson plan text and the usage attribute parameters are input into a random forest model for training, the lesson plan text is classified according to the usage attributes, and the set of lesson plan texts with the same usage attributes forms the lesson plan set.

[0012] As a preferred embodiment of the technical solution of the present invention, the corresponding teaching object ratio, teaching object priority, scheduling priority, and teaching environment influencing factors are obtained according to the identity information, the teaching scheduling information, the teaching object information, the teaching objective information and the teaching scene information; the teaching object parameters include teaching object information, teaching object ratio, and teaching object priority; the teaching scene parameters include scheduling priority and teaching environment influencing factors.

[0013] As a preferred embodiment of the technical solution of the present invention, the teaching plan generation step specifically includes: A simplification sub-step, combining and integrating the material keywords and the extended content in the teaching plan text according to the knowledge point keywords to form the simplification teaching plan text; The integration sub-step calculates the logical difference of the knowledge point relationship in each node of the different simplified lesson plan texts. When the logical difference is within a preset range, the corresponding simplified lesson plan texts are integrated into the total lesson plan text. When the logical difference exceeds the preset range, the corresponding theoretical keywords, the material keywords and the content corresponding to the material keywords in the corresponding simplified lesson plan texts are added into the total lesson plan text according to the knowledge point relationship.

[0014] As a preferred embodiment of the technical solution of the present invention, according to the priority relationship of the theoretical keywords contained in the total lesson plan text, the total lesson plan text is updated to the single keyword total lesson plan library corresponding to the theoretical keyword with the highest priority, and the update priority score is calculated later according to the usage of the total lesson plan text. Each time the single keyword total lesson plan library is updated, the total lesson plan text contained therein is prioritized.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Generate demand commands based on the lesson plan requirements input by teachers, perform targeted content extraction, extract keywords from the three dimensions of the text - theoretical dimension, material dimension, and template dimension, classify attributes, form usable data for big data, extract relationship logic, and generate knowledge point maps. After learning through the big data model, further optimize the lesson plan to obtain a general lesson plan that conforms to the teaching process to cope with different temporary needs in the classroom, and provide teaching paths and case content when encountering various temporary situations.

[0016] 2. Data training is performed through the random forest model, and the category prediction is performed for each input lesson plan text. The prediction results are given. The corresponding usage attributes of each lesson plan text are predicted, and all lesson plan texts are classified according to the usage attributes to obtain a set of lesson plans with different usage attribute classifications, which can be effectively used in subsequent analysis applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a schematic diagram of an online page of a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph in the present invention.

[0018] Figure 2 The present invention is a flow chart of a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph.

[0019] Figure 3 This is a schematic diagram of an example of a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph in the present invention.

[0020] Figure 4 This is a schematic diagram of an example of a method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph in the present invention. DETAILED DESCRIPTION

[0021] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0022] At present, the development of educational big models mainly adopts two technical routes: one is to directly call the general big model and make it have certain professional capabilities through fine-tuning or prompt learning. The other is to use professional data in the field of education to train a big model specifically for solving educational tasks. However, although both technical routes have made certain progress, the implementation effect still needs to be improved. The problem is that due to the lack of sufficient professional data training and insufficient in-depth knowledge in the field of education, the current big model is not very intelligent and it is difficult to flexibly handle complex and changeable educational tasks. The solution to developing educational big models lies in integrating the two technical routes. This is not a simple addition, but the use of application-driven, co-construction and sharing methods to continuously obtain data from normalized educational applications through open data interfaces, realize the organic combination of "big model" and "small model", "big data" and "small data", meet the actual needs of teachers and students in daily teaching, and break the "data island". At the same time, with the expert knowledge base as a supplement to the big model, the educational knowledge and teaching methods are consciously "taught" to the big model, and various intelligent educational technologies are integrated and applied to form a special big model with the ability to flexibly handle various complex educational tasks.

[0023] A method for automatically generating electronic teaching plans of knowledge points based on a knowledge graph comprises the following steps: The extraction step receives the demand command including the teaching demand content and identity information, extracts the theoretical keywords and material keywords in the demand content text, associates the theoretical keywords with the material keywords, forms a relationship set, and obtains a knowledge relationship map.

[0024] The extraction steps specifically include: The receiving sub-step receives a demand command including teaching demand content and identity information, which may be a direct demand command generated by importing text in a fixed demand template format, or a demand command in a demand content form generated after receiving input data.

[0025] In the classification sub-step, the demand content text is extracted. The identity information is the fixed information carried by the teaching demander who sends the demand command, which is bound to the teaching demander and obtained by querying the identity ID. The demand content text is the instruction content of the demand command itself. According to different attribute tags, the demand content text is classified by attributes to obtain the theoretical content text, material content text and template text.

[0026] The keyword extraction sub-step identifies the theoretical content text, material content text and template text information, and obtains the corresponding theoretical keywords, material keywords and template numbers respectively.

[0027] In the graph construction sub-step, theoretical keywords and material keywords are associated, keyword entities and knowledge point relationships between theoretical keywords and material keywords are extracted, and a knowledge relationship graph is generated.

[0028] For example, the theoretical keyword is Romanticism; the corresponding material keywords are Schubert, Brahms, Liszt, and Beethoven, generating a knowledge relationship map such as Figure 3 shown.

[0029] Analyze the feature vector of the material keyword based on the theoretical keyword, match the material supplementary material according to the feature vector query, fill the material supplementary material into the material keyword, generate branch theoretical keywords of the theoretical keywords, associate the branch theoretical keywords with the material keywords, and update the knowledge relationship map.

[0030] Teachers log in to their own accounts on their own terminals and upload the demand form directly with their own accounts or enter the teaching plan requirements online, as shown in the teaching plan demand form and Figure 1 If the teaching plan requirement form is directly submitted, the requirement command is generated directly through OCR recognition. If it is filled out online, the requirement command is generated based on the multiple filled-in contents received.

[0031] Theory 1 Theory 2 Material 1 Material 2 Material 3 Requirement template number Table 1. Teaching plan requirement form Among them, theoretical knowledge points are filled in theory 1, theory 2, etc., which are generally text data, and may be image data in a few cases. For example, "Romanticism" is filled in theory 1, and the theoretical keyword extracted through text is "Romanticism" in the later stage; material content is filled in material 1, material 2, material 3, etc. Each column is the relevant material corresponding to the theoretical keyword in the same column, and each row represents the number of related materials corresponding to the theoretical keyword. The material type is generally unformatted data, which can be pictures, text, URLs, video and audio. The material needs to be extracted unformatted, and text information can be extracted through OCR recognition. For example, the material keywords extracted from the corresponding theoretical keywords are "Chopin", "Brahms", and "Rachmaninoff". The template number of the requirement is obtained according to the filled-in requirement template number, such as M1 and M2. Finally, the requirement command is generated.

[0032] Since teachers log in through their own accounts, when submitting, they are accompanied by identity ID information, personal identity information of the teacher under the teacher identity (including teaching subject information), teacher's teaching schedule information, teaching object information corresponding to each class, teaching purpose information and teaching scene information (such as teaching classroom type, scale, etc.). According to the classification and keywords, a usage format containing keywords and classification is formed, such as "teaching plan for elementary school students on understanding musical romanticism". The format includes at least one or more fields of "teaching object", "teaching purpose" and "teaching scene".

[0033] In the matching step, the knowledge relationship map is filled into the matching lesson plan template to form at least one lesson plan outline, and the extended content matching the material keywords is searched, and the extended content is added into the lesson plan outline to form the lesson plan text.

[0034] The matching steps specifically include: In the matching sub-step, the lesson plan template is retrieved from the lesson plan template library according to the template number.

[0035] Generate sub-steps and fill the knowledge relationship map into the lesson plan template to form a lesson plan outline.

[0036] The expansion sub-step searches for expansion content that matches the material keyword, and adds the expansion content to the lesson plan outline to form the lesson plan text. For example, if the material keyword is "Brahms", the expansion content can include "Personal Profile: Johannes Brahms (1833-1897) is a German composer. He is considered the last composer of German classicism and one of the representatives of mid-Romanticism. Brahms is an important figure in the history of music in the 19th century. His works build a bridge between classicism and romanticism, and are famous for their superb skills and profound emotional expression.", "Music Style: Brahms's musical style combines the rigorous structure of classicism. and romantic emotional expression, forming a unique artistic feature. Specifically, it includes the combination of classical form and romantic emotion, the lyricism of melody, the richness of harmony, the use of counterpoint and polyphony, the diversity of rhythm, and the rich colors of orchestral music. "," Major works: Symphony No. 1 (Op. 68), Ein Deutsches Requiem (Op. 45), Hungarian Dances (WoO1), Violin Concerto in D major (Violin Concerto in D major) Oncerto in D Major, Op. 77), Symphony No. 3 in F Major, Op. 90, Symphony No. 4 in E Minor, Op. 98, Lullaby No. 4 in Wiegenlied, Op. 49, Piano Quartet No. 2 in A Major, Op. 26, Social Influence: Johannes Brahms is An important composer of the Romantic music period in the 19th century, his musical works and personal experiences have had a profound impact on society. Some critics rank him together with Bach and Beethoven as the "Three Bs". While inheriting the tradition of classical music, his works also show the characteristics of Romantic music, such as the lyricism of melody, the richness of harmony, the use of counterpoint and polyphony, the diversity of rhythm and the rich colors of orchestral music. Brahms' musical legacy is rich and far-reaching. His works are novel and distinctive in melody, harmony and texture, and have had a profound influence on later musicians."

[0037] All lesson plan template entities are identified to obtain the module entities in each module text in the template. The process logic relationship between the module entities is extracted based on the context relationship, and the module entities are compared with the keyword entities for similarity to obtain the most suitable lesson plan template.

[0038] The training step generates usage attribute parameters based on identity information. The usage attribute parameters include at least teaching object parameters, teaching purpose parameters and teaching scenario parameters. The lesson plan text is input into the large model for training and classification to obtain a lesson plan set with different usage attribute classifications.

[0039] During the training step: The usage attribute parameters are generated according to the identity information, the lesson plan text and the usage attribute parameters are input into the random forest model for training, the lesson plan text is classified according to the usage attributes, and the lesson plan text set with the same usage attributes forms a lesson plan set.

[0040] According to the identity information, teaching schedule information, teaching object information, teaching purpose information and teaching scenario information, the corresponding teaching object ratio, teaching object priority, scheduling priority and teaching environment influencing factors are obtained. Teaching object parameters include teaching object information, teaching object ratio and teaching object priority. Teaching scenario parameters include scheduling priority and teaching environment influencing factors.

[0041] The lesson plan generation step is to merge the content of the lesson plan texts in the lesson plan set to form a simplified lesson plan text, and unify and integrate multiple simplified lesson plan texts. Each simplified lesson plan text belongs to a different lesson plan set, and branch lesson plan content is formed at different teaching logical relationships, forming a total lesson plan text with branch lesson plan content and branch teaching paths. For example: analyze the logical relationship between "Chopin", "Brahms", "Rachmaninoff" and "Romanticism" respectively. "Chopin" is a famous representative of the "Romanticism" and belongs to the early period of the Romanticism. The logical relationship between "Chopin" and "Romanticism" is "Chopin is an early representative of the Romanticism". Similarly, the logical relationship between "Brahms" and "Romanticism" is "Brahms is a middle representative of the Romanticism", and the logical relationship between "Rachmaninoff" and "Romanticism" is "Rachmaninoff is a late representative of the Romanticism". According to the logic, the branch theory keywords can be extracted after "Romanticism": "early", "middle", "late", and the specific knowledge point map is as follows Figure 4 shown.

[0042] The demand command includes identity attribute information and demand content text. The identity attribute information includes identity ID, identity information, teaching schedule information, teaching object information, teaching purpose information and teaching scenario information. The demand content text includes theoretical content, material content and template content.

[0043] The steps of lesson plan generation include: In the simplification sub-step, the material keywords and the extended content in the lesson plan text are merged and integrated according to the knowledge point keywords to form a simplification lesson plan text. For example, the extended content is formed into expandable (the content is expanded according to the label when used) secondary labels, such as "personal profile", "music style", "main works", and "social impact".

[0044] The integration sub-step calculates the logical differences in the relationships between knowledge points in each node of different condensed lesson plan texts. When the logical differences are within a preset range, the corresponding condensed lesson plan texts are integrated into a total lesson plan text. When the logical differences exceed the preset range, the corresponding theoretical keywords, material keywords, and the contents corresponding to the material keywords in the corresponding condensed lesson plan texts are added to the total lesson plan text according to the relationships between knowledge points.

[0045] Example 2 Based on Example 1, the difference from Example 1 is that: According to the priority relationship of the theoretical keywords contained in the total lesson plan text, the total lesson plan text is updated to the single keyword total lesson plan library corresponding to the theoretical keyword with the highest priority. Later, the update priority score is calculated based on the usage of the total lesson plan text. After each update of the single keyword total lesson plan library, the total lesson plan texts contained therein are prioritized.

[0046] The usage of the total lesson plan text by teachers is obtained, which is divided into offline teaching scenarios and online teaching scenarios. For online teaching scenarios, the material keywords used in the total lesson plan are extracted according to the usage data of the teacher in class, and the theoretical keywords to which the material keywords used are assigned points according to the time used. For example, the theoretical keyword with the longest time in a single course is assigned 10 points, the second longest theoretical keyword is assigned 9 points, and so on. Then the theoretical keyword with the highest score is selected, and the total lesson plan text after use is stored in the keyword total lesson plan library of the corresponding keyword. The total lesson plan text in the keyword total lesson plan library is basically scored 100 points. The subsequent teacher can also directly search the existing total lesson plan text in the keyword total lesson plan library for lesson preparation. If the total lesson plan text is used, the total lesson plan text is assigned +1; if the total lesson plan text is not used after being searched and browsed, the total lesson plan text is assigned -1; the keyword total lesson plan library is updated every time it is used or a new total lesson plan text enters.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution 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 solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

Claims

1. A method for automatically generating electronic teaching plans of knowledge points based on knowledge graph, characterized in that: The following steps are involved: An extraction step, receiving a demand command including teaching demand content and identity information, extracting theoretical keywords and material keywords in the demand content text, associating the theoretical keywords with the material keywords to form a relationship set, and obtaining a knowledge relationship map; A matching step, filling the knowledge relationship map into a matching teaching plan template to form at least one teaching plan outline, searching for extended content matching the material keyword, and adding the extended content into the teaching plan outline to form a teaching plan text; A training step, generating usage attribute parameters based on the identity information, wherein the usage attribute parameters at least include a teaching object parameter, a teaching purpose parameter, and a teaching scenario parameter, and inputting the teaching plan text into a large model for training and classification to obtain a teaching plan set with different usage attribute classifications; The lesson plan generation step is to merge the content of the lesson plan texts in the lesson plan set to form a simplified lesson plan text, unify and integrate multiple simplified lesson plan texts, form branch lesson plan contents at different points of teaching logic relationships, and form a total lesson plan text with the branch lesson plan contents and branch teaching paths.

2. According to claim 1, a method for automatically generating electronic teaching plans of knowledge points based on knowledge graphs is characterized in that: The demand command includes identity attribute information and the demand content text, wherein the identity attribute information includes identity ID, identity information, teaching schedule information, teaching object information, teaching purpose information and teaching scenario information; the demand content text includes theoretical content, material content and template content.

3. According to claim 2, a method for automatically generating electronic teaching plans of knowledge points based on knowledge graphs is characterized in that: The extraction step specifically includes: A receiving sub-step, receiving the demand command including teaching demand content and identity information; The classification sub-step is to extract the demand content text, classify the demand content text by attributes according to different attribute tags, and obtain the theoretical content text, the material content text and the template text; The keyword extraction sub-step is to identify the theoretical content text, the material content text and the template text information, and obtain the theoretical keywords, the material keywords and the template number respectively; The graph construction sub-step associates the theoretical keywords with the material keywords, extracts the keyword entities and knowledge point relationships between the theoretical keywords and the material keywords, and generates the knowledge relationship graph.

4. According to claim 3, a method for automatically generating electronic teaching plans of knowledge points based on knowledge graphs is characterized in that: Analyze the feature vector of the material keyword based on the theoretical keyword, query and match the material supplementary material according to the feature vector, fill the material supplementary material into the material keyword, generate branch theoretical keywords of the theoretical keywords, associate the branch theoretical keywords with the material keywords, and update the knowledge relationship map.

5. According to claim 3, a method for automatically generating electronic teaching plans of knowledge points based on knowledge graphs is characterized in that: The matching step specifically includes: A matching sub-step, retrieving a corresponding teaching plan template from a teaching plan template library according to the template number; A generation sub-step, filling the knowledge relationship map into the teaching plan template to form the teaching plan outline; The expansion sub-step is to search for expansion contents matching the material keywords, and add the expansion contents into the teaching plan outline to form the teaching plan text.

6. The method for automatically generating electronic teaching plans of knowledge points based on knowledge graph according to claim 5, characterized in that: All the lesson plan template entities are identified to obtain the module entities in each module text in the template, the process logic relationship between the module entities is extracted according to the context relationship, and the module entities are compared with the keyword entities for similarity to obtain the most suitable lesson plan template.

7. The method for automatically generating electronic teaching plans of knowledge points based on knowledge graph according to claim 3 is characterized in that: In the training step: The usage attribute parameters are generated according to the identity information, the lesson plan text and the usage attribute parameters are input into a random forest model for training, the lesson plan text is classified according to the usage attributes, and the set of lesson plan texts with the same usage attributes forms the lesson plan set.

8. The method for automatically generating electronic teaching plans of knowledge points based on knowledge graph according to claim 7, characterized in that: According to the identity information, the teaching scheduling information, the teaching object information, the teaching objective information and the teaching scene information, the corresponding teaching object ratio, teaching object priority, scheduling priority and teaching environment influencing factors are obtained; the teaching object parameters include teaching object information, teaching object ratio and teaching object priority; the teaching scene parameters include scheduling priority and teaching environment influencing factors.

9. The method for automatically generating electronic teaching plans of knowledge points based on knowledge graph according to claim 8, characterized in that: The teaching plan generation step specifically includes: A simplification sub-step, combining and integrating the material keywords and the extended content in the teaching plan text according to the knowledge point keywords to form the simplification teaching plan text; The integration sub-step calculates the logical difference of the knowledge point relationship in each node of the different simplified lesson plan texts. When the logical difference is within a preset range, the corresponding simplified lesson plan texts are integrated into the total lesson plan text. When the logical difference exceeds the preset range, the corresponding theoretical keywords, the material keywords and the content corresponding to the material keywords in the corresponding simplified lesson plan texts are added into the total lesson plan text according to the knowledge point relationship.

10. The method for automatically generating electronic teaching plans of knowledge points based on knowledge graph according to claim 1, characterized in that: According to the priority relationship of the theoretical keywords contained in the total lesson plan text, the total lesson plan text is updated to the single keyword total lesson plan library corresponding to the theoretical keyword with the highest priority. Later, the update priority score is calculated according to the usage of the total lesson plan text. Each time the single keyword total lesson plan library is updated, the total lesson plan text contained therein is prioritized.

Images