Intelligent teaching material analysis system based on internal and external resources

Abstract

The application provides a teaching material intelligent analysis system and platform based on internal resources and external resources, which is used for constructing a teaching material knowledge graph according to digital teaching materials, open data of external Internet and type tags based on the open data, and has the characteristics of including a teaching material preprocessing module, a teaching material structure analysis module, an internal resource linking module, an external resource linking module and a teaching material knowledge graph module. In summary, the method can construct a teaching material knowledge graph with high-quality internal and external resources, and construct a platform for providing more abundant digital teaching material related information.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of intelligent education, and particularly relates to a teaching material intelligent analysis system and platform based on internal and external resources. BACKGROUND

[0002] At present, paper-based teaching materials still account for the majority in the field of education compared with other types of teaching materials, but paper-based teaching materials have the problems of difficult retrieval and expansion. With the popularity of mobile phones, computers and other devices, electronic reading has gradually become an important reading way in people's learning and life, and the digitization of teaching materials has also begun to attract people's attention and gradually become a current research hotspot, and more and more digital teaching materials have appeared on the market.

[0003] At present, the main forms of digital teaching materials include static media digital teaching materials, multimedia digital teaching materials and rich media digital teaching materials. The static media digital teaching materials emphasize the digital reduction of paper-based teaching materials, and only store the paper-based teaching materials in the form of digitalization in the electronic device, which is easy to retrieve but lacks interactivity. The multimedia digital teaching materials organize related resources in the form of web pages or directly insert multimedia resources in the digital teaching materials, which can effectively integrate and present various resources, but there is a lack of association between these resources, and learners are still passive in accepting resource information. The rich media digital teaching materials pay more attention to user experience, and provide more rich user interaction on the basis of multimedia digital teaching materials, so that learners change from passive acceptance to active learning, but the cost of constructing rich media digital teaching materials is high, and the teaching material resources lack systematic knowledge guidance.

[0004] The current digital teaching material platform makes digital management and display of paper-based teaching materials, but the utilization rate of the resources in the teaching materials is not high, and the knowledge system and ability system behind the resources have not been formed, so it is impossible to guide intelligent learning across disciplines from all perspectives. This problem mainly reflects in the following points: first, the educational resources in the teaching materials are in a fragmented form and scattered everywhere, and various resources exist in isolation, and there is a lack of association between resources; second, the resource sources in the teaching materials are closed, and only contain the internal resources possessed by the teaching materials, without considering the large amount of high-quality resources existing in the external Internet; third, the resources in the teaching materials lack systematic knowledge guidance, and the resources are essentially the embodiment of subject knowledge, and the resources must be associated with the corresponding subject knowledge or subject concept to play a role in application. SUMMARY

[0005] The present application is made to solve the above problems, and aims to provide a teaching material intelligent analysis system and platform based on internal and external resources.

[0006] The application provides a teaching material intelligent analysis system based on internal resources and external resources, which is used for constructing a teaching material knowledge graph according to digital teaching materials, open data of external Internet and type tags based on the open data, and has the characteristics that the system comprises a teaching material preprocessing module, a teaching material structure analysis module, an internal resource linking module, an external resource linking module and a teaching material knowledge graph module. The teaching material preprocessing module comprises a teaching material source file conversion submodule, a teaching material page arrangement submodule and a teaching material character recognition submodule. The teaching material source file conversion submodule is used for converting digital teaching materials in different file forms into digital teaching materials in a unified format for searching. The teaching material page arrangement submodule is used for uniformly arranging the pages of the teaching materials in the unified format to obtain formatted teaching materials. The teaching material character recognition submodule is used for recognizing the content of the formatted teaching materials through OCR character recognition to obtain pure text content in the content as teaching material analysis text. The teaching material structure analysis module comprises a term mining submodule, a key relationship and attribute extraction submodule and an internal element mining submodule. The term mining submodule is used for mining the teaching material analysis text to obtain multiple terms in the teaching material analysis text as domain terms. The key relationship and attribute extraction submodule is used for extracting words from the teaching material analysis text as the relationship between two domain terms or the attribute of a domain term through a reading comprehension method. A relationship triple is formed by the domain terms and the corresponding relationship, and an attribute triple is formed by two domain terms and the corresponding attribute. The internal element mining submodule is used for dividing the content of the formatted teaching materials into different regions through a document layout analysis tool, and dividing the different regions into text resources containing only text content and complex structure data containing other content. The internal resource linking module is used for performing OCR character recognition processing on the text resources according to the position information of the text resources in the formatted teaching materials, linking the text resources to related domain terms as text resource triples, and linking the complex structure data to related domain terms as complex structure data triples according to the position information of the complex structure data in the formatted teaching materials. The external resource linking module comprises a term and resource type matching submodule, an external resource discovery submodule and an external resource mounting submodule. The term and resource type matching submodule is used for classifying the domain terms according to type tags, and matching each domain term to a corresponding type tag. The external resource discovery submodule is used for searching open data through a search engine according to the type tags corresponding to the domain terms to obtain open data related to the domain terms and the type tags as related open data. The external resource mounting submodule is used for further filtering and screening the related open data, linking the filtered and screened related open data to the domain terms as open data triples. The teaching material knowledge graph module is used for constructing a teaching material knowledge graph according to the relationship triples, the attribute triples, the text resource triples, the complex structure data triples and the open data triples.

[0007] In the internal resource and external resource-based teaching material intelligent analysis system provided by the application, the teaching material text recognition submodule can further have the following features: the teaching material analysis text is a TXT format file with position information.

[0008] In the internal resource and external resource-based teaching material intelligent analysis system provided by the application, the internal resource linking module can further have the following features: the specific process of linking the text resource to the related domain term is as follows: performing OCR text recognition processing on the text resource, fusing the text resources in different regions according to the OCR text recognition processing result and the position information, performing text understanding and classification on the fusion result, obtaining the corresponding domain term, and linking the text resource corresponding to the fusion result to the domain term.

[0009] In the internal resource and external resource-based teaching material intelligent analysis system provided by the application, the internal resource linking module can further have the following features: the specific process of linking the complex structure data to the related domain term is as follows: obtaining the title and context information of the complex structure data according to the position information of the complex structure data, matching the title with the domain term, if the matching is successful, linking the complex structure term to the matched domain term, if the matching fails, counting the frequency of occurrence of the term vocabulary in the context information, and linking the complex structure term to the domain term corresponding to the term vocabulary with the highest frequency of occurrence.

[0010] In the internal resource and external resource-based teaching material intelligent analysis system provided by the application, the external resource mounting submodule can further have the following features: the specific process of filtering and screening the related open data is as follows: uniformly representing the related open data in different modalities through a multi-modal representation and alignment algorithm, filtering the related open data according to a double-tower sorting algorithm and sorting them from high to low according to quality, and selecting the first n related open data in the sequence as the filtered and screened related open data.

[0011] The application further provides a digital teaching material platform, comprising a front-end submodule, a back-end submodule and a database, wherein the front-end submodule is used to display digital teaching materials online using a reader, send an information reading request to the back-end submodule according to a selected term in the digital teaching materials, and display the related information of the term fed back by the back-end submodule; the back-end submodule is used to read the related information of the domain term corresponding to the term in the teaching material knowledge graph from the database according to the information reading request and feed back to the front-end submodule; and the database is used to store the data of the teaching material knowledge graph, which is constructed by inputting the digital teaching materials into the internal resource and external resource-based teaching material intelligent analysis system according to any one of the above.

[0012] In the digital teaching material platform provided by the application, the reader can be a PDF reader.

[0013] Effects of the application

[0014] According to the teaching material intelligent analysis system and platform based on internal resources and external resources, important resources and fragmented resources in the digital teaching material are mined and associated through the internal resource linking module, and on the other hand, high-quality resources are screened from the external resources and associated with the corresponding digital teaching material through the external resource linking module, so as to deeply associate and integrate multi-source and multi-modal resources with subject knowledge, and further expand the knowledge capacity of the digital teaching material. Therefore, the teaching material intelligent analysis system and platform based on internal resources and external resources can construct a high-quality internal and external resource teaching material knowledge graph and provide more rich digital teaching material related information to platform users. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 is a frame diagram of the teaching material intelligent analysis system in the embodiment of the application;

[0016] Figure 2 is a frame diagram of the teaching material preprocessing module in the embodiment of the application;

[0017] Figure 3 is a frame diagram of the teaching material structure analysis module in the embodiment of the application;

[0018] Figure 4 is a frame diagram of the external resource linking module in the embodiment of the application;

[0019] Figure 5 is a flow diagram of constructing a teaching material knowledge graph in the embodiment of the application;

[0020] Figure 6 is a frame diagram of the digital teaching material platform in the embodiment of the application;

[0021] Figure 7 is a diagram of the image display area of the front-end sub-module in the embodiment of the application;

[0022] Figure 8 is a flow diagram of displaying related information of a term in the embodiment of the application. DETAILED DESCRIPTION

[0023] In order to make the technical means, creative features, purposes and effects achieved by the application easy to understand, the following embodiments will be specifically described in combination with the drawings.

[0024] Figure 1is a framework schematic diagram of the teaching material intelligent analysis system in the embodiment of the present application.

[0025] As shown in Figure 1 , the teaching material intelligent analysis system 100 is used to construct a teaching material knowledge graph according to digital teaching materials, open data of external Internet and type tags based on the open data, and includes a teaching material preprocessing module 10, a teaching material structure analysis module 20, an internal resource linking module 30, an external resource linking module 40 and a teaching material knowledge graph module 50.

[0026] Figure 2 is a framework schematic diagram of the teaching material preprocessing module in the embodiment of the present application.

[0027] As shown in Figure 2 , the teaching material preprocessing module 10 includes a teaching material source file conversion sub-module 101, a teaching material page layout sub-module 102 and a teaching material text recognition sub-module 103. The teaching material preprocessing module 10 is used to process teaching materials in multiple different file forms into machine-processable teaching materials in the same format for subsequent modules and algorithms to process uniformly. For digital teaching material files input in any form such as PDF, Word, etc., the teaching material preprocessing module 10 converts the digital teaching material files into TXT format files with position information and stores them in corresponding database tables.

[0028] The teaching material source file conversion sub-module 101 is used to convert digital teaching materials in different file forms into unified format digital teaching materials for searching as unified format teaching materials. In this embodiment, the unified format teaching materials are PDF version digital teaching materials, which are convenient for searching and querying fine-grained content such as highlighting and identifying terms in the digital teaching materials when the digital teaching materials are displayed on the digital teaching material platform.

[0029] The teaching material page layout sub-module 102 is used to uniformly layout the pages of the unified format teaching materials to obtain formatted teaching materials. In this embodiment, the teaching material page layout sub-module 102 eliminates formats such as headers and footers that have a negative impact on automatic modules such as text recognition, which is convenient for page display of the digital teaching material platform 200.

[0030] The teaching material text recognition sub-module 103 is used to recognize the content of the formatted teaching materials through OCR text recognition to obtain pure text content in the content as teaching material analysis text for further processing of the content by subsequent natural language processing methods such as term mining.

[0031] The teaching material analysis text is a TXT format file with position information.

[0032] Figure 3 is a framework schematic diagram of the teaching material structure analysis module in the embodiment of the present application.

[0033] As shown in Figure 3As shown, the textbook structure analysis module 20 includes a term mining submodule 201, a key relationship and attribute extraction submodule 202, and an internal element mining submodule 203. The textbook structure analysis module 20 is configured to analyze important resources and structures within a digital textbook, and provide a data basis for internal resource linking and external resource linking.

[0034] The term mining submodule 201 is configured to mine a textbook analysis text to obtain a plurality of terms in the textbook analysis text as domain terms. In this embodiment, the domain terms output are representative terms of the corresponding domain of the textbook analysis text. For example, if the input is a machine learning textbook, domain terms such as “linear model” and “decision tree” are mined, thereby helping other modules to focus on high-quality domain terms and integrate and link related high-quality resources.

[0035] The key relationship and attribute extraction submodule 202 is configured to extract words from the textbook analysis text as relationships between two domain terms, including hierarchical relationships, predecessor-successor relationships, and the like, or attributes of domain terms, such as abbreviations, English names, and the like, to form relationship triples composed of domain terms and corresponding relationships, and attribute triples composed of two domain terms and corresponding attributes. In this embodiment, the key relationship and attribute extraction submodule labels relevant tail entities and attribute values on corresponding relationships or attributes for each domain term input, and obtains relationship triples or attribute triples after rule post-processing.

[0036] The internal element mining submodule 203 is configured to divide the content of a formatted textbook into different regions by a document layout analysis tool, and divide the different regions into text resources containing only text content and complex structure data containing other content. In this embodiment, the different regions include text, pictures, picture titles, tables, table titles, references, formulas, and the like, the text resources include text, picture titles, table titles, and the like, and the complex structure data includes pictures, formulas, tables, and the like.

[0037] The internal resource linking module 30 is configured to perform OCR character recognition processing on text resources according to position information of the text resources in a formatted textbook, link the text resources to related domain terms as text resource triples, and link complex structure data to related domain terms as complex structure data triples according to position information of the complex structure data in the formatted textbook.

[0038] Specifically, the process of linking text resources to related domain terms includes the following steps:

[0039] The text resource is subjected to OCR character recognition processing, the text resources in different regions are fused according to the OCR character recognition processing result and position information, the fusion result is subjected to text understanding and classification, the corresponding domain term is obtained, and the text resource corresponding to the fusion result is linked to the domain term.

[0040] The specific process of fusing the text resources in different regions is as follows:

[0041] The text resource is input into an OCR engine, and the output is a bounding box and a text string of a certain line of text, the fusion of the contents in different regions is performed in the manner of calculating the ratio of the intersection area of the OCR bounding box and each region to the area of the OCR bounding box, if the ratio is greater than a certain threshold value, it is considered that the bounding box belongs to the region, and then the upper and lower position relationship is obtained according to the position information, and the OCR results are merged line by line.

[0042] The specific process of linking the complex structure data to the related domain term is as follows:

[0043] According to the position information of the complex structure data, the title and context information of the complex structure data are obtained, the title is matched with the domain term, if the matching is successful, the complex structure term is linked to the matched domain term, if the matching fails, the appearance frequency of the term vocabulary in the context information is counted, and the complex structure term is linked to the domain term corresponding to the term vocabulary with the highest appearance frequency.

[0044] Figure 4 It is a framework diagram of the external resource linking module in the embodiment of the application.

[0045] As shown in Figure 4 , the external resource linking module 40 includes a term and resource type matching sub-module 401, an external resource discovery sub-module 402, and an external resource mounting sub-module 403.

[0046] The term and resource type matching sub-module 401 is used for classifying the domain terms according to the type tags and matching each domain term with the corresponding type tag. In the embodiment, the type tag is a human tag.

[0047] The external resource discovery sub-module 402 is used for searching the open data according to the designed query algorithm through a search engine according to the type tag corresponding to the domain term, obtaining the open data related to the domain term and the type tag as related open data. In the embodiment, the open data comes from multiple channels such as encyclopedias, forums, and databases, and is expressed in multiple types such as videos, codes, papers, and news.

[0048] The external resource mounting sub-module 403 is used for further filtering and screening the related open data, linking the filtered and screened related open data to the domain term as open data triples.

[0049] The textbook knowledge graph module 50 is configured to construct a textbook knowledge graph according to the relation triplets, the attribute triplets, the text resource triplets, the complex structure data triplets and the open data triplets.

[0050] Figure 5 FIG. 1 is a schematic diagram of a process for constructing a textbook knowledge graph according to an embodiment of the present application.

[0051] As shown in FIG. 1, the process for constructing a textbook knowledge graph includes the following steps: Figure 5

[0052] Step S1, the textbook source file conversion sub-module 101 converts digital textbooks in different file formats into digital textbooks in a unified format for searching, as unified format textbooks.

[0053] Step S2, the textbook page arrangement sub-module 102 arranges the pages of the unified format textbooks uniformly, to obtain formatted textbooks.

[0054] Step S3, the textbook text recognition sub-module 103 recognizes the content of the formatted textbooks through OCR text recognition, to obtain pure text content in the content as textbook analysis text.

[0055] Step S4, the term mining sub-module 201 mines the textbook analysis text, to obtain a plurality of terms in the textbook analysis text as domain terms.

[0056] Step S5, the key relation and attribute extraction sub-module 202 extracts words from the textbook analysis text as relations or attributes through a reading comprehension method, and forms relation triplets from the domain terms and corresponding relations, and forms attribute triplets from two domain terms and corresponding attributes.

[0057] Step S6, the internal element mining sub-module 203 divides the content of the formatted textbooks into different regions through a document layout analysis tool, and divides the different regions into text resources containing only text content and complex structure data containing other content.

[0058] Step S7, the internal resource linking module 30 performs OCR text recognition processing on the text resources according to the position information of the text resources in the formatted textbooks, links the text resources to related domain terms as text resource triplets, and links the complex structure data to related domain terms as complex structure data triplets according to the position information of the complex structure data in the formatted textbooks.

[0059] Step S8, the term and resource type matching sub-module 401 classifies the domain terms according to type tags, and matches each domain term to a corresponding type tag.

[0060] ​In step S9, the external resource discovery submodule 402 retrieves open data related to the domain terminology and type tag by using a search engine and a pre-designed query algorithm based on the type tag corresponding to the domain terminology, and uses this as relevant open data.

[0061] In step S10, the external resource mounting submodule 403 further filters and selects the relevant open data, and links the filtered and selected relevant open data to domain terms as open data triples.

[0062] In step S11, the textbook knowledge graph module 50 constructs the textbook knowledge graph based on relation triples, attribute triples, text resource triples, complex structure data triples, and open data triples.

[0063] Figure 6 This is a schematic diagram of the framework of the digital teaching material platform in an embodiment of the present invention.

[0064] like Figure 6 As shown, the digital teaching material platform 200 includes a front-end sub-module 60, a back-end sub-module 70, and a database 80.

[0065] The front-end submodule 60 is used to display digital teaching materials online using a reader. Based on selected terms in the digital teaching materials, it sends information retrieval requests to the back-end submodule 70 and displays the relevant information of the terms returned by the back-end submodule 70. In this embodiment, a web application front-end is used to display the digital teaching materials using a PDF reader. The reader's built-in annotation function enables interaction between the electronic teaching materials and the web application. Users trigger events by clicking, and the web application sends requests to the back-end submodule based on the clicked annotation content, displaying the relevant feedback information.

[0066] The rear sub-module 70 is used to read relevant information of the domain terminology corresponding to the terminology in the textbook knowledge graph from the database 80 according to the information reading request, and feed it back to the front sub-module 60.

[0067] Database 80 is used to store data of the textbook knowledge graph, which is constructed by inputting digital textbooks into the textbook intelligent analysis system 100.

[0068] Figure 7 This is a schematic diagram of the image display area of ​​the front-end module in an embodiment of the present invention.

[0069] like Figure 7As shown, the image display area of the front-end sub-module 60 of the digital teaching material platform 200 includes a digital teaching material display area 601 and a related information display area 602. The digital teaching material display area 601 is used to display the digital teaching material in the digital teaching material platform 200 in the form of pdf. The related information display area 602 is used to display the related information of the terms in the digital teaching material. After the user clicks on the term text in the digital teaching material display area 601, the related information display area 602 displays a plurality of related information 603 of the term text.

[0070] Figure 8 is a flowchart of displaying the related information of the term in the embodiment of the present application.

[0071] Step T1, the front-end sub-module 60 displays the digital teaching material in the digital teaching material display area 601 online using the reader.

[0072] Step T2, the user selects a term in the digital teaching material.

[0073] Step T3, the front-end sub-module 60 sends an information reading request to the back-end sub-module 70 according to the selected term.

[0074] Step T4, the back-end sub-module 70 reads the related information of the domain term corresponding to the term in the teaching knowledge graph from the database 80 according to the information reading request, and feeds back to the front-end sub-module 60.

[0075] Step T5, the front-end sub-module 60 displays the related information of the term fed back by the back-end sub-module 70 in the related information display area 602.

[0076] Effects of the embodiment

[0077] According to the teaching material intelligent analysis system and platform based on internal and external resources related to the present embodiment, on the one hand, the internal resource linking module is used to mine important resources and fragmented resources in the digital teaching material and associate them, and on the other hand, the external resource linking module is used to associate the high-quality resources from the external resources with the corresponding digital teaching material, so as to deeply associate and integrate the multi-source and multi-modal resources with the subject knowledge, and further expand the knowledge capacity of the digital teaching material. In summary, the present method can construct a teaching knowledge graph with high-quality internal and external resources, and the present platform can provide more rich related information of the digital teaching material.

[0078] The above-mentioned embodiments are preferred cases of the present application and do not limit the protection scope of the present application.

Claims

1. A teaching material intelligent analysis system based on internal resources and external resources, for constructing a teaching material knowledge graph according to digital teaching materials, open data of external Internet and type tags based on the open data, characterized in that, The method comprises the following steps: The teaching material preprocessing module, the teaching material structure analysis module, the internal resource linking module, the external resource linking module and the teaching material knowledge graph module, The teaching material preprocessing module comprises a teaching material source file conversion submodule, a teaching material page arrangement submodule and a teaching material text recognition submodule, The teaching material source file conversion submodule is configured to convert the digital teaching materials in different file formats into digital teaching materials in a unified format for searching, as unified format teaching materials, The teaching material page arrangement submodule is configured to uniformly arrange the pages of the unified format teaching materials to obtain formatted teaching materials, The teaching material text recognition submodule is configured to recognize the content of the formatted teaching materials by OCR text recognition to obtain pure text content in the content as teaching material analysis text, The teaching material structure analysis module comprises a term mining submodule, a key relationship and attribute extraction submodule and an internal element mining submodule, The term mining submodule is configured to mine the teaching material analysis text to obtain a plurality of terms in the teaching material analysis text as domain terms, The key relationship and attribute extraction submodule is configured to extract words from the teaching material analysis text as a relationship between two domain terms or an attribute of a domain term by a reading comprehension method, to form a relationship triple consisting of the domain term and the corresponding relationship, and to form an attribute triple consisting of two domain terms and the corresponding attribute, The internal element mining submodule is configured to divide the content of the formatted teaching materials into different regions by a document layout analysis tool, and to divide the different regions into text resources containing only text content and complex structure data containing other content, The internal resource linking module is configured to perform OCR text recognition processing on the text resources according to the location information of the text resources in the formatted teaching materials, link the text resources to related domain terms as text resource triples, and link the complex structure data to related domain terms as complex structure data triples according to the location information of the complex structure data in the formatted teaching materials, The external resource linking module comprises a term and resource type matching submodule, an external resource discovery submodule and an external resource mounting submodule, The term and resource type matching submodule is configured to classify the domain terms according to the type tags, and match each domain term with a corresponding type tag, The external resource discovery submodule is configured to retrieve the open data related to the domain terms and the type tags as related open data by a search engine according to the type tags corresponding to the domain terms, The external resource mounting submodule is configured to further filter and screen the related open data, and link the filtered and screened related open data to the domain terms as open data triples, The textbook knowledge graph module is configured to construct the textbook knowledge graph based on the relation triplets, the attribute triplets, the text resource triplets, the complex structure data triplets and the open data triplets.

2. The intelligent textbook analysis system based on internal resources and external resources according to claim 1, characterized in that: wherein In the textbook character recognition submodule, the textbook analysis text is a TXT format file with position information.

3. The intelligent textbook analysis system based on internal resources and external resources according to claim 1, characterized in that: wherein In the internal resource linking module, the specific process of linking the text resource to the related domain term is as follows: The text resource is subjected to OCR character recognition processing, the text resource in different regions is fused based on the OCR character recognition processing result and the position information, the fused result is subjected to text understanding and classification, the corresponding domain term is obtained, and the text resource corresponding to the fused result is linked to the domain term.

4. The intelligent textbook analysis system based on internal resources and external resources according to claim 1, characterized in that: wherein In the internal resource linking module, the specific process of linking the complex structure data to the related domain term is as follows: The title and context information of the complex structure data are obtained based on the position information of the complex structure data, the title is matched with the domain term, if the matching is successful, the complex structure term is linked to the matched domain term, if the matching fails, the appearance frequency of the term vocabulary in the context information is counted, and the complex structure term is linked to the domain term corresponding to the term vocabulary with the highest appearance frequency.

5. The intelligent textbook analysis system based on internal resources and external resources according to claim 1, characterized in that: wherein, In the external resource mounting submodule, the specific process of filtering and screening the related open data is as follows: The related open data in different modalities are uniformly represented by a multi-modal representation and alignment algorithm, the related open data are filtered and sorted from high to low in quality according to a double-tower sorting algorithm, and the first n related open data in the sequence are selected as the filtered and screened related open data.

6. A smart teaching platform, characterized in that, It comprises: a front-end submodule, a back-end submodule and a database, The front-end submodule is configured to display a digital textbook online using a reader, send an information reading request to the back-end submodule according to a selected term in the digital textbook, and display the related information of the term fed back by the back-end submodule. The back-end submodule is configured to read the related information of the domain term corresponding to the term in the textbook knowledge graph from the database according to the information reading request, and feed back to the front-end submodule. The database is configured to store the data of the textbook knowledge graph. The textbook knowledge graph is constructed by inputting the digital textbook into the intelligent textbook analysis system based on internal resources and external resources according to any one of claims 1 to 5.

7. The smart teaching platform of claim 6, wherein: wherein the reader is a pdf reader.

Images