A knowledge graph construction method, device, equipment and product

Abstract

The application provides a knowledge graph construction method, device, equipment and product, the method comprises the following steps: extracting the directory in the teaching material, obtaining a plurality of groups of knowledge point hierarchical directories, wherein each group of knowledge point hierarchical directory comprises at least one directory; the last level directory under each group of knowledge point hierarchical directory is matched with the body text in the teaching material, and the body text content corresponding to the last level directory under each group of knowledge point hierarchical directory is obtained; based on the entity and entity relationship extraction of the body text content corresponding to the last level directory under each group of knowledge point hierarchical directory, a teaching knowledge graph is constructed. The application can improve the accuracy and efficiency of constructing the knowledge graph.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a knowledge graph construction method, device, equipment and product. Background Art

[0002] Knowledge graphs can integrate massive amounts of dispersed information into a structured, interconnected network, thereby improving the efficiency and accuracy of information retrieval, data analysis, and intelligent decision-making. Therefore, building high-quality knowledge graphs is of great significance in multiple fields.

[0003] Currently, the construction of knowledge graphs primarily relies on manual annotation and organization by domain experts. For example, in the field of education, experts typically identify and define key knowledge points and concepts based on curriculum guidelines, textbook content, and their own teaching experience, and then organize the relationships between them to construct the knowledge graph structure. However, this manual approach often suffers from inaccuracies and low efficiency. Summary of the Invention

[0004] Based on the above-mentioned technical status, this application provides a knowledge graph construction method, device, equipment and product, which can improve the accuracy and efficiency of knowledge graph construction.

[0005] In order to achieve the above technical objectives, this application specifically proposes the following technical solutions:

[0006] According to the first aspect of the embodiment of the present application, a knowledge graph construction method is provided, including: extracting the directory in the teaching materials to obtain multiple groups of knowledge point hierarchical directories, wherein each group of knowledge point hierarchical directories includes at least one first-level directory; matching the last-level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories; extracting entities and entity relationships based on the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories, to construct a teaching knowledge graph.

[0007] In some implementations, the method of extracting the catalogue from the teaching materials to obtain multiple groups of hierarchical catalogues of knowledge points includes: identifying a first keyword and a second keyword in the teaching materials, wherein the first keyword represents the starting position of the catalogue and the second keyword represents the ending position of the catalogue; extracting each title between the first keyword and the second keyword; and dividing each title into directories of corresponding levels according to the format of each title and the preset format corresponding to each level of catalogue in each group of hierarchical catalogues of knowledge points to obtain the multiple groups of hierarchical catalogues of knowledge points.

[0008] In some implementations, the text includes various text fragments, and the matching of the last-level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories includes: performing the following steps for each group of knowledge point hierarchical directories: determining the similarity between the last-level directory under the group of knowledge point hierarchical directories and the various text fragments; determining the text fragment with the highest similarity to the last-level directory under the group of knowledge point hierarchical directories as the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories; or merging at least one text fragment whose similarity to the last-level directory under the group of knowledge point hierarchical directories is greater than or equal to a first similarity threshold to obtain the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories.

[0009] In some implementations, after the text segment with the highest similarity to the last-level directory under the group of knowledge point hierarchical directories is determined as the content corresponding to the last-level directory under the group of knowledge point hierarchical directories, the method also includes: if the similarity between the first text segment and the last-level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the first text segment and the last-level directory under the group of knowledge point hierarchical directories, then generating an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories; wherein the similarity between the first text segment and the last-level directory under the group of knowledge point hierarchical directories is greater than or equal to a second similarity threshold.

[0010] In some implementations, after merging at least one text segment whose similarity with the last level directory under the group of knowledge point hierarchical directories is greater than or equal to the first similarity threshold to obtain the text content corresponding to the last level directory under the group of knowledge point hierarchical directories, the method further includes: if the text segment corresponding to the similarity among the various similarities is less than the first similarity threshold and greater than or equal to the third similarity threshold, if the similarity between the second text segment and the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories, then generating an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories; wherein the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories is less than the first similarity threshold and greater than or equal to the third similarity threshold.

[0011] In some implementations, each text segment contains key elements, and the key elements include at least one of a title, a table, a picture, a paragraph, and an annotation; wherein, determining the similarity between the last level directory under the group of knowledge point hierarchical directories and the respective text segments includes: determining the similarity between the last level directory under the group of knowledge point hierarchical directories and each key element in each text segment; and performing weighted summation of the similarities between the last level directory under the group of knowledge point hierarchical directories and each key element in each text segment according to the weight coefficients corresponding to the respective key elements, to obtain the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment.

[0012] In some implementations, the teaching materials include multiple teaching materials, the teaching knowledge graph includes a teaching knowledge graph corresponding to each teaching material, and the method further includes: identifying entities representing the same object and relationships representing the same semantics in multiple teaching knowledge graphs; associating the entities representing the same object in the multiple teaching knowledge graphs, and associating the relationships representing the same semantics to obtain a comprehensive teaching knowledge graph.

[0013] According to the second aspect of the embodiment of the present application, a knowledge graph construction device is provided, including: an extraction unit, used to extract the directory in the teaching materials to obtain multiple groups of knowledge point hierarchical directories, wherein each group of knowledge point hierarchical directories includes at least one first-level directory; a matching unit, used to match the last-level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories; a construction unit, used to extract entities and entity relationships based on the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories, and construct a teaching knowledge graph.

[0014] According to the third aspect of an embodiment of the present application, an electronic device is provided, comprising a memory and a processor; the memory is connected to the processor for storing programs; the processor is used to implement the knowledge graph construction method as described in the first aspect by running the program in the memory.

[0015] According to a fourth aspect of an embodiment of the present application, a computer program product is provided, comprising computer program instructions, which, when executed by a processor, cause the processor to execute: the knowledge graph construction method as described in the first aspect.

[0016] The embodiments of the present application provide a method, device, equipment and product for constructing a knowledge graph, which extracts a directory from teaching materials to obtain a hierarchical directory of knowledge points including at least one level of directory, and matches the corresponding text content in the teaching materials with respect to the last level of directory under each group of knowledge point hierarchical directories. Subsequently, entities and relationships are extracted from the text content corresponding to the last level of directory under each group of knowledge point hierarchical directories, and finally a structured teaching knowledge graph is constructed. This technical solution is based on the existing directory system structure in the teaching materials as a knowledge hierarchy framework, which enables accurate positioning and content matching of the text content, thereby providing a structured foundation for subsequent knowledge extraction and graph construction. Compared with the traditional method of relying on manual organization of knowledge structure, it can improve the accuracy and context relevance of knowledge extraction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0018] Figure 1 A flowchart of a knowledge graph construction method provided in an embodiment of the present application.

[0019] Figure 2 A flowchart of a directory extraction process provided in an embodiment of the present application.

[0020] Figure 3 A flowchart of a directory and text matching process provided in an embodiment of the present application.

[0021] Figure 4 A flowchart of another directory and text matching process provided in an embodiment of the present application.

[0022] Figure 5 A flowchart for determining similarity based on structured text provided in an embodiment of the present application.

[0023] Figure 6 A schematic diagram of the structured processing provided in an embodiment of the present application.

[0024] Figure 7 A schematic diagram of the structure of a knowledge graph construction device provided in an embodiment of the present application.

[0025] Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] The technical solutions proposed in the embodiments of this application are applicable to any application scenarios in the field of education, such as auxiliary teaching scenarios, personalized learning scenarios, intelligent question-answering scenarios, knowledge management and discovery scenarios, and educational research scenarios. Specifically:

[0027] In assisted teaching scenarios, knowledge graphs can help teachers efficiently organize and manage teaching resources and assist in developing teaching plans and scheduling. For example, by combining the syllabus with students' current knowledge, relevant knowledge points, teaching cases, and exercises can be automatically recommended, improving teaching quality and efficiency.

[0028] In personalized learning scenarios, the knowledge graph can recommend personalized learning paths and resources based on students' individual learning situations and interests, helping them learn and review efficiently. For example, it can recommend appropriate course content, exercises, and additional reading materials based on their level.

[0029] In intelligent question-and-answer scenarios, knowledge graphs, combined with natural language processing technology, can provide intelligent question-and-answer services, quickly responding to education-related questions raised by students or teachers. For example, when a student asks, "What is Newton's second law?", detailed explanations and practical application examples can be provided using the entities and relationships in the knowledge graph.

[0030] In the context of knowledge management and discovery, knowledge graphs help systematically organize and store educational knowledge, facilitating knowledge retrieval, sharing, and dissemination. Furthermore, by analyzing the structure and relationships within knowledge graphs, new knowledge connections and patterns can be discovered, promoting innovation in educational content.

[0031] In educational research scenarios, it can provide educational researchers with rich data support and analytical tools, facilitating research on educational theory, teaching methods, and subject knowledge structure. For example, researchers can use knowledge graphs to analyze the impact of different teaching strategies on students' mastery of knowledge points, or explore the inherent connections between knowledge across disciplines, thereby promoting the development of educational practice and theory.

[0032] The technical solutions provided in the embodiments of the present application can be exemplarily applied to hardware devices such as processors, electronic devices, and servers (including cloud servers), or packaged into software programs and run. When the hardware devices execute the processing of the technical solutions in the embodiments of the present application, or the above-mentioned software programs are run, the automatic splitting of target tasks and the automatic calling of the application program interfaces required for the tasks can be achieved, thereby completing the purpose of the target tasks. The embodiments of the present application only provide an illustrative introduction to the specific processing of the technical solutions in the present application, and do not limit the specific implementation form of the technical solutions in the present application. Any technical implementation form that can execute the processing of the technical solutions in the present application can be adopted by the embodiments of the present application.

[0033] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0034] Before introducing this application solution, we first introduce the relevant technologies:

[0035] A knowledge graph is a structured semantic knowledge base that represents real-world entities (such as people, places, objects, and concepts) and their relationships in a graphical form. A knowledge graph typically consists of nodes and edges, with nodes representing entities and edges representing the various semantic relationships between entities. This highly organized approach not only captures and expresses complex knowledge structures but also helps machines better understand natural language and perform complex reasoning and learning tasks. This plays a vital role in a variety of fields, including search engine optimization, recommendation systems, natural language processing, and intelligent question-answering.

[0036] Specifically, applying knowledge graphs to search engines can provide more accurate and relevant search results by understanding the contextual meaning of user queries; applying knowledge graphs to recommendation systems can provide users with personalized content recommendations based on user preferences and behavior patterns; applying knowledge graphs to natural language processing can support machines' understanding and analysis of the deep semantics of language and improve text processing capabilities; applying knowledge graphs to intelligent question-answering systems can utilize the structured knowledge in the graph to achieve fast and accurate answers to questions; applying knowledge graphs to data integration can integrate data from different sources into a structured knowledge framework, improving data consistency and availability.

[0037] The educational knowledge graph is a structured knowledge representation method built for the field of education. Its core is to organize and store knowledge points, concepts, resources and other contents in the field of education in the form of a graph, so that it has good interpretability and operability, and is easy for machines to recognize and process, thus providing a basis for promoting the informatization and intelligence of education.

[0038] However, the current construction of educational knowledge graphs mainly relies on manual annotation and organization by domain experts, which leads to the problems of insufficient accuracy and low efficiency of the constructed knowledge graphs.

[0039] In view of this, the embodiments of the present application are dedicated to providing a knowledge graph construction method, device, equipment and product. Specifically, the method obtains multiple groups of knowledge point hierarchical directories by extracting directories from teaching materials. Subsequently, the last level directory under each group of knowledge point hierarchical directories is matched with the text in the teaching materials, thereby realizing the association between each group of knowledge point hierarchical directories and the text part. Finally, based on these associated information, entities and their relationships are further extracted to construct an accurate teaching knowledge graph, which can not only improve the accuracy of the constructed knowledge graph, but also greatly improve the construction efficiency. Detailed description is given one by one in the following embodiments.

[0040] Exemplary Methods

[0041] Figure 1 This is a flow chart of a knowledge graph construction method provided in an embodiment of the present application. Figure 1 As shown, the knowledge graph construction method provided in this embodiment includes steps S101-S103:

[0042] S101. Extract the catalogue from the teaching materials to obtain multiple groups of knowledge point hierarchical catalogues, wherein each group of knowledge point hierarchical catalogues includes at least one level catalogue.

[0043] In this embodiment, the teaching materials include textbooks, course outlines, academic papers, and various teaching-related materials, etc. These teaching materials contain knowledge points, concepts, terms, and their relationships in various subject areas, forming rich knowledge resources.

[0044] A table of contents is an overview of the various sections of a textbook. It helps you quickly understand the overall structure and easily locate the sections of interest. Tables of contents are usually organized by chapter or topic, indicating the title of each section and the corresponding page number or location.

[0045] When building a knowledge graph, a knowledge point hierarchy can be constructed based on the catalog of teaching materials (such as textbooks and course outlines). This knowledge point hierarchy not only summarizes the content framework of the teaching materials but also demonstrates the logical connections between knowledge points through hierarchical relationships, facilitating further analysis and processing of these knowledge points and their relationships.

[0046] Each set of knowledge point hierarchical directories may include at least one first-level directory. Taking any set of knowledge point hierarchical directories as an example, it may include only a first-level directory, or may include a first-level directory and its subordinate multiple subdirectories, such as a second-level directory, or further include a second-level directory and a third-level directory.

[0047] The specific implementation method of extracting the directory from the above teaching materials can be found below Figure 2 A detailed description of the embodiments shown in:

[0048] Figure 2 A flowchart of a catalog extraction process is provided for an embodiment of the present application. As shown in FIG. 1, step S101 extracts a catalog in teaching materials to obtain a plurality of sets of knowledge point hierarchical catalogs, which specifically includes the following sub-steps S201-S203: Figure 2

[0049] S201, identifies a first keyword and a second keyword in the teaching materials, the first keyword representing a starting position of the catalog, and the second keyword representing an ending position of the catalog.

[0050] In a specific implementation, an OCR (Optical Character Recognition) technology can be used to recognize the text content in the teaching materials, so as to locate the starting position identifier and the ending position identifier of the catalog.

[0051] The first keyword can be, for example, the word "catalog", and the second keyword can be, for example, the word "reference file", "appendix", "answer", or "acknowledgement".

[0052] When the keyword "catalog" is recognized, it is determined as the starting point of the catalog, and when the keywords "reference file", "appendix", "answer", and "acknowledgement" are recognized, it is determined as the ending point of the catalog.

[0053] S202, extracts each title located between the first keyword and the second keyword.

[0054] In this step, the OCR technology is continued to be used to recognize and extract each title between the starting position of the catalog and the ending position of the catalog, to provide data support for subsequent generation of a structured catalog.

[0055] S203, divides each title to a corresponding hierarchical catalog according to the format of each title and a preset format corresponding to each hierarchical catalog in each set of knowledge point hierarchical catalogs, to obtain the plurality of sets of knowledge point hierarchical catalogs.

[0056] Since different levels of catalogs usually have different format characteristics (such as font, font size, numbering, indentation method, etc.), the format of the extracted title can be compared with the preset format corresponding to each hierarchical catalog in each set of knowledge point hierarchical catalogs, to determine the catalog level to which each title belongs.

[0057] For example, the preset format corresponding to each hierarchical catalog in the knowledge point hierarchical catalog is as follows:

[0058] ​First-level headings should be in Arial font, size 16, bold, and unnumbered; second-level headings should be in Arial font, size 14, bold, and numbered with Roman numerals (I, II, III...); third-level headings should be in Times New Roman font, size 12, not bold, and numbered with Arabic numerals (1, 2, 3...) and parentheses.

[0059] Suppose Heading A is in Arial font, size 16, bold, and unnumbered; Heading B is in Arial font, size 14, bold, and numbered with the Roman numeral "I"; and Heading C is in Times New Roman font, size 12, non-bold, and numbered with "1). Heading A conforms to the default format for a first-level heading, Heading B conforms to the default format for a second-level heading, and Heading C conforms to the default format for a third-level heading. Therefore, Heading A, Heading B, and Heading C are first-level, second-level, and third-level headings, respectively.

[0060] Through the above method, each title can be classified into the corresponding directory level to obtain multiple groups of knowledge point hierarchical directories.

[0061] The above steps S201-S203 introduce the specific implementation method of generating a directory from the teaching materials. However, in some embodiments, the directory in the teaching materials may also exist in the form of an independent directory file. For this situation, the application also provides the following processing method:

[0062] For example, before executing step S201, first determine whether there is an independent directory file in the teaching materials; if there is an independent directory file, directly read the directory file to obtain multiple sets of knowledge point hierarchical directories; if there is no independent directory file, execute the extraction process of steps S201-S203 to obtain multiple sets of knowledge point hierarchical directories.

[0063] After obtaining multiple sets of knowledge point classification catalogs from teaching materials, you can associate the multiple sets of knowledge point classification catalogs with the main text content in the teaching materials to facilitate subsequent knowledge extraction. For the specific association process, please refer to Figure 1 , the following is a detailed description of step S102:

[0064] S102. Match the last level directory under each group of knowledge point hierarchical directories with the main text in the teaching materials to obtain the main text content corresponding to the last level directory under each group of knowledge point hierarchical directories.

[0065] By segmenting the main text content in the teaching materials through the knowledge point hierarchical catalog, it can be divided into parts corresponding to each group of knowledge point hierarchical catalogs, so that in the subsequent knowledge extraction process, it is possible to focus on text fragments related to specific knowledge points, avoid interference from irrelevant information, and thus improve the accuracy and relevance of knowledge extraction.

[0066] In some embodiments, vector matching technology can be used to achieve semantic-level matching between the hierarchical catalog of knowledge points and the text segments. For example, by vectorizing the catalog and the text separately and using semantic similarity calculation methods, the text content most semantically relevant to each catalog can be accurately identified. This method not only captures the surface matching relationships between texts, but also deeply explores their underlying semantic connections, thereby more accurately locating the text content corresponding to the catalog.

[0067] Figure 3 This is a flowchart of a directory and text matching process provided by an embodiment of the present application. Figure 3 As shown, when executing step S102, it specifically includes: executing the following sub-steps S301-S302 for each group of knowledge point classification catalogs:

[0068] S301: Determine the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment.

[0069] In some embodiments, step S301 includes: vectorizing the last level directory under the hierarchical directory of the group of knowledge points to obtain a corresponding directory vector; vectorizing each text segment in the teaching materials to obtain a text segment vector corresponding to each text segment; calculating the similarity between the directory vector and each text segment vector to obtain the similarity between the last level directory under the hierarchical directory of the group of knowledge points and each text segment.

[0070] The similarity between the catalog vector and each text segment vector can be measured using similarity calculation methods such as cosine similarity, Euclidean distance, Manhattan distance, Pearson correlation coefficient, and Jaccard similarity.

[0071] Alternatively, a deep learning-based vector matching model (such as a pre-trained language model such as BERT) can be used to generate semantic vectors for the last-level directory and each text segment, and perform similarity calculations.

[0072] S302: Determine the text segment with the highest similarity to the last level directory under the group of knowledge point hierarchical directories as the text content corresponding to the last level directory under the group of knowledge point hierarchical directories.

[0073] After obtaining the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment, the text segments that are greater than or equal to the second similarity threshold can be sorted in order of similarity from high to low, and the text segment corresponding to the highest similarity can be determined as the text content corresponding to the last level directory under the group of knowledge point hierarchical directories.

[0074] In some other embodiments, steps S301 and S302 can also be implemented through a large language model, that is, the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment is calculated through a large language model, and the text segment with the highest similarity to the last level directory under the group of knowledge point hierarchical directories is determined as the text content corresponding to the last level directory under the group of knowledge point hierarchical directories, as the final output result of the model.

[0075] In order to further improve the accuracy of knowledge extraction, after associating multiple groups of knowledge point hierarchical directories with the text content through the above steps S301 and S302, this embodiment can also provide the construction of association relationships between knowledge point hierarchical directories across groups. Figure 3 After step S302, the following step S303 may be further included:

[0076] S303. If the similarity between the first text segment and the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories, then generate an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories.

[0077] The similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories is greater than or equal to a second similarity threshold.

[0078] In this step, the first text segment can be understood as at least one text segment having a similarity greater than or equal to a second similarity threshold, excluding the highest similarity, in the similarity ranking results of the last level directory under the group of knowledge point hierarchical directories.

[0079] If the similarity between the first text segment and the last level directory under the knowledge point hierarchical directory of this group is greater than or equal to the second similarity threshold, and the similarity between the first text segment and the last level directory of a certain other group of knowledge point hierarchical directory is higher than its similarity under the original group directory, then it is considered that the text segment is more suitable for the knowledge point system structure of the other group, and its correlation with the original group directory is higher.

[0080] For ease of understanding, step S303 is further explained below with reference to a specific example:

[0081] In some examples, suppose knowledge extraction is performed on text segment A, resulting in entity A. Entity A has a similarity of 0.62 with section 1 and 0.78 with section 2. If the second similarity threshold is set to 0.5, then because the similarity between entity A and section 2 exceeds the threshold, a triple is created: "Section 2 contains entity A." Furthermore, because the similarity between entity A and section 1 also exceeds the threshold, entity A is linked to section 1 across categories to reflect the potential semantic connection between the two knowledge points.

[0082] In other examples, suppose knowledge extraction is performed on text segment A, resulting in entities B and C. The similarity between entity B and section 1 is 0.63, and the similarity between entity C and section 2 is 0.71. 0.5 is still used as the second similarity threshold. In this case, the triples "Section 1 contains entity B" and "Section 2 contains entity C" are generated, respectively. A cross-segment association is then established between entities B and C, reflecting the possible cross-knowledge point semantic connections between different entities.

[0083] Through the above method, not only can the structured mapping between knowledge points and entities be achieved, but also the cross-level and cross-segment associations between entities can be effectively explored, thereby improving the integrity and semantic expression ability of the constructed teaching knowledge graph, and facilitating subsequent cross-topic knowledge recommendation, auxiliary learning path planning and other application scenarios.

[0084] Figure 4 This is a flowchart of another directory and text matching process provided by the embodiment of this application. Figure 4 As shown, when executing step S102, it specifically includes: executing the following sub-steps S401-S402 for each group of knowledge point classification catalogs:

[0085] S401: Determine the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment.

[0086] The specific implementation of step S401 is the same as that of the aforementioned step S301. For details, please refer to the detailed introduction of the specific implementation of step S301, which will not be repeated here.

[0087] S402: Merge at least one text segment having a similarity greater than or equal to a first similarity threshold with the last level directory under the group of knowledge point hierarchical directories to obtain text content corresponding to the last level directory under the group of knowledge point hierarchical directories.

[0088] After obtaining the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment, the text segments greater than or equal to the second similarity threshold can be sorted in descending order of similarity, and at least one text segment with a similarity greater than or equal to the first similarity threshold is taken, and the at least one merged text segment is determined to be the text content corresponding to the last level directory under the group of knowledge point hierarchical directories.

[0089] The first similarity threshold is greater than or equal to the second similarity threshold.

[0090] After associating multiple groups of knowledge point hierarchical directories with the text content through the above steps S401 and S402, in order to further improve the accuracy of knowledge extraction, this embodiment can also provide an implementation method for cross-directory association. Figure 4 After step S402, the following step S403 may be further included:

[0091] S403. If, among the text segments corresponding to the similarities that are less than the first similarity threshold and greater than or equal to the third similarity threshold, there exists a second text segment whose similarity with the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories, then an association relationship is generated between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories.

[0092] The first similarity threshold is greater than the third similarity threshold. The similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories is less than the first similarity threshold and greater than or equal to the third similarity threshold.

[0093] In this step, the second text segment can be understood as at least one text segment in the similarity ranking result of the last level directory under the group of knowledge point hierarchical directories, whose similarity with the last level directory under the group of knowledge point hierarchical directories is less than the first similarity threshold, and greater than or equal to the third similarity threshold, and whose similarity with the last level directory under other groups of knowledge point hierarchical directories is greater than the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories.

[0094] That is to say, if the similarity between the second text segment and the last level directory under the knowledge point hierarchical directory of this group is less than the first similarity threshold and greater than or equal to the third similarity threshold, and the similarity between the second text segment and the last level directory of another group of knowledge point hierarchical directory is higher than its similarity under the original group directory, then it is considered that the text segment is more suitable for the knowledge point system structure of the other group, and its correlation with the original group directory is higher.

[0095] Therefore, a cross-group association record can be generated, which includes the directory identifiers of the source and target groups, the text segment IDs, and the similarity value between the two. These associations can be used in subsequent application scenarios such as knowledge graph construction, cross-theme knowledge recommendation, and auxiliary learning path planning.

[0096] Continue reading Figure 1 , after step S102, step S103 may also be included.

[0097] S103. Based on the extraction of entities and entity relationships from the text content corresponding to the last level directory under each group of knowledge point hierarchical directory, a teaching knowledge graph is constructed.

[0098] In some embodiments, the specific implementation method of step S103 includes: preprocessing the text content corresponding to the last level directory under each group of knowledge point hierarchical directory to obtain preprocessed text content; performing entity recognition and relationship extraction based on the preprocessed text content to obtain multiple triples; and constructing a teaching knowledge graph based on multiple triples.

[0099] Among them, preprocessing includes: dividing the text content corresponding to the last level directory under each group of knowledge point hierarchical directory into words or vocabulary units, and removing high-frequency words that are not helpful for understanding the text content, such as "的" and "是", and assigning grammatical roles to each word, such as noun or verb.

[0100] When performing entity recognition and relationship extraction, rule-based methods, supervised learning methods, or deep learning methods can be used. Among them, rule-based methods refer to matching entities and their relationships in text through pre-defined grammatical rules and patterns, for example, by writing regular expressions or using dependency parsing to analyze specific relationship patterns.

[0101] Supervised learning methods require a well-labeled training dataset containing labeled entities and their relationships. Machine learning algorithms (such as support vector machines (SVMs) and decision trees) are then used to train the model to automatically identify entity relationships within the text.

[0102] Deep learning methods also require a well-labeled training dataset containing labeled entities and their relationships. Deep learning algorithms, such as convolutional neural networks (CNNs), recurrent neural networks (RNNs) and their variants (such as LSTMs and GRUs), and Transformer architectures (such as BERT), are then used to train neural network models to automatically identify entity relationships within textual content.

[0103] After obtaining entities and their relationships, we can use the entities as nodes in the knowledge graph, and the relationships as connections between nodes. That is, if a relationship exists between two entities, a line will be drawn between the nodes corresponding to the two entities; otherwise, no line will be drawn. This is used to indicate whether a relationship exists between two entities.

[0104] This embodiment adopts a content matching mechanism based on directory guidance, that is, it uses the existing directory system in the teaching materials as the skeleton of the knowledge hierarchy to locate and match the corresponding text content, and uses this as the basis for subsequent knowledge extraction and graph construction. Compared with the traditional method of relying on manual organization of knowledge structure, it can improve the accuracy and efficiency of knowledge graph construction. In addition, by extracting the entities and their relationships of the text content corresponding to the last level directory under each group of knowledge point hierarchical directory, the constructed teaching knowledge graph not only has rich semantic information, but also can accurately reflect the logical hierarchy and association relationship between knowledge points, so as to better fit the knowledge organization and retrieval needs in the teaching scenario.

[0105] In order to more accurately measure the semantic relevance between the knowledge point hierarchical directory and the main text content, this embodiment can also adopt a structured processing mechanism for teaching materials. That is, by identifying and extracting key elements in the main text content and performing weighted calculations based on their importance in knowledge expression, the accuracy and relevance of similarity judgments can be improved. Specifically, when determining the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment based on structured text, the following embodiment can be provided:

[0106] Figure 5 The flowchart of determining similarity based on structured text provided in the embodiment of this application. Figure 5 As shown, when determining the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment, step S301 may further include the following steps S501 and S502:

[0107] S501: Determine the similarity between the last level directory under the group of knowledge point hierarchical directories and each key element in each text segment.

[0108] Figure 6 This is a schematic diagram of the structured processing provided in the embodiment of the present application. Figure 6 As shown, by conducting a structural analysis of the text content in the teaching materials, key elements are identified and extracted, including at least one of the following: titles, tables, pictures, paragraphs, and annotations. Annotations include comments, footnotes, and key points.

[0109] After that, the last level directory under the set of knowledge points hierarchical directory and each type of key element in each text segment are respectively vectorized to obtain the directory vector and the key element vector corresponding to each type of key element in each text segment. Traditional word vector models such as TF-IDF, Word2Vec, GloVe, or pre-trained language models based on deep learning (such as BERT, RoBERTa, ERNIE, etc.) can be used to vectorize each type of key element. For non-text key elements (such as pictures), image recognition technology can be used to extract semantic descriptions or use multi-modal models for vector encoding.

[0110] Further, for each text segment, the similarity between the directory vector and the key element vector corresponding to each key element in the text segment is calculated to obtain the similarity between the last level directory under the set of knowledge points hierarchical directory and each key element in each text segment.

[0111] The similarity between the directory vector and each key element can be measured using similarity calculation methods such as cosine similarity (CosineSimilarity), Euclidean distance (Euclidean Distance), Manhattan distance (Manhattan Distance), Pearson correlation coefficient (Pearson Correlation Coefficient), and Jaccard similarity (JaccardSimilarity).

[0112] Alternatively, an end-to-end semantic matching model (such as BERT-based Sentence Pair Classification) can also be used to directly output the similarity between the two.

[0113] S502, according to the weight coefficient corresponding to each key element, the similarity between the last level directory under the set of knowledge points hierarchical directory and each key element in each text segment is weighted and summed to obtain the similarity between the last level directory under the set of knowledge points hierarchical directory and each text segment.

[0114] After obtaining the similarity between each key element in each text segment and the knowledge point directory, the weight coefficient corresponding to each key element can be used to reflect the relative importance of different key elements in knowledge expression. The similarity of each key element is weighted and summed to obtain the final similarity, which is as follows:

[0115] Sim(final)=w1*Sim(title)+w2*Sim(paragraph)+w3*Sim(table)+w4*Sim(image) + w5*Sim(annotation);

[0116] Among them, w1 ~ w5 are the weight coefficients corresponding to each key element, and they satisfy: w1+w2+ w3+w4+w5=1.

[0117] In actual applications, the weight coefficient can be dynamically adjusted according to the type of teaching materials (such as textbooks, handouts, PPTs, etc.), subject areas (such as liberal arts focusing on paragraphs, science may focus on charts) and user feedback, and the weight distribution can also be automatically optimized through machine learning.

[0118] In some examples, titles are often highly summarized, highly condensed, and highly indicative of the core content of the text. Their value range can be set to 0.25-0.35. The body paragraphs carry the main knowledge information and are the core source of knowledge extraction. Their value range can be set to 0.20-0.30. Tables are used to structure data or compare information and are of moderate importance. Their value range can be set to 0.10-0.20. Images are often used to aid understanding, and their semantic importance needs to be judged in conjunction with the caption or context. Their value range can be set to 0.05-0.15. Annotations such as highlights and comments provide supplementary explanations, but have a lower information density. Their value range can be set to 0.05-0.10.

[0119] It should be noted that the weight allocation method introduced in the above example is only for illustrative purposes and is not intended to limit the specific allocation method of weight coefficients in this application. In actual application, the corresponding weight parameters can be flexibly set according to specific business needs, application scenarios or data characteristics.

[0120] This embodiment structures the text content into multiple key elements and assigns them different semantic weights, which can more accurately represent the semantic association relationship between the knowledge point directory and the text fragments, thereby improving the accuracy of knowledge extraction.

[0121] After processing various teaching materials through the above-mentioned embodiments to obtain a series of independent teaching knowledge graphs, in order to achieve unified representation and interoperability of knowledge and further improve the integration and utilization efficiency of educational resources, these knowledge graphs generated based on different teaching materials can also be merged and completed to generate a comprehensive knowledge graph.

[0122] Among them, the merging of knowledge graphs includes: identifying entities representing the same object in multiple teaching knowledge graphs, and relationships representing the same semantics; associating entities representing the same object in multiple teaching knowledge graphs, and associating relationships representing the same semantics to obtain a comprehensive teaching knowledge graph.

[0123] In this embodiment, it is first necessary to identify entities representing the same object in multiple teaching knowledge graphs. Specifically, it includes: comparing and analyzing the names, description information and association relationships of entities in different graphs. For example, if two graphs respectively contain the knowledge point of "Newton's Second Law", it is necessary to determine whether the two knowledge points represent the same concept and standardize them (such as unified naming conventions) to ensure the consistency of concepts. Among them, natural language processing technology, such as text similarity calculation or semantic analysis, can be used to determine whether the two knowledge points point to the same concept.

[0124] The relationships between entities also need to be identified and matched. For example, the mathematical expressions for "Newton's Second Law" and its relationships with "force," "mass," and "acceleration" may have different representations or levels of detail in different knowledge graphs. By comparing and analyzing the specific content of these relationships, we ensure they are correctly understood and connected in the new comprehensive graph, thereby ensuring the consistency and coherence of the knowledge system.

[0125] After identifying the entities and their relationships that represent the same object, it is necessary to extract related entities and relationships from the original graph and integrate them into the comprehensive teaching knowledge graph. This involves creating new nodes and edges or adjusting the existing structure to reflect a more comprehensive knowledge system.

[0126] In order to further improve the integrity of the knowledge graph, after completing the knowledge graph merging operation, you can also perform the knowledge graph completion operation.

[0127] Specifically, if it is identified that there is a lack of direct connection between certain entities in the merged graph, but based on existing knowledge it is determined that such a connection should exist, these "missing links" can be automatically filled in through reasoning algorithms to enhance the coherence and depth of the knowledge graph.

[0128] In addition, authoritative data sources from other related fields can be introduced as supplements, such as academic databases, encyclopedias, etc., to add more dimensions of information to the comprehensive knowledge graph.

[0129] In addition, user feedback channels can be established to collect actual usage experience and suggestions provided by users to help discover and correct potential problems, so that the comprehensive knowledge graph can be more in line with educational needs and improve its applicability and user satisfaction.

[0130] Exemplary devices

[0131] Corresponding to the above-mentioned knowledge graph construction method, an embodiment of the present application also provides a knowledge graph construction device. Figure 7 This is a schematic diagram of the structure of a knowledge graph construction device provided in an embodiment of the present application. Figure 7 As shown, the knowledge graph construction device provided by the embodiment of the present application includes: an extraction unit 701, a matching unit 702 and a construction unit 703; wherein the extraction unit 701 is used to extract the directory in the teaching material to obtain multiple groups of knowledge point hierarchical directories, wherein each group of knowledge point hierarchical directories includes at least one level directory; the matching unit 702 is used to match the last level directory under each group of knowledge point hierarchical directories with the text in the teaching material to obtain the text content corresponding to the last level directory under each group of knowledge point hierarchical directories; the construction unit 703 is used to extract entities and entity relationships based on the text content corresponding to the last level directory under each group of knowledge point hierarchical directories to construct a teaching knowledge graph.

[0132] In some embodiments, when the extraction unit 701 extracts the directory in the teaching materials and obtains multiple groups of knowledge point hierarchical directories, it specifically includes: identifying the first keyword and the second keyword in the teaching materials, the first keyword represents the starting position of the directory, and the second keyword represents the end position of the directory; extracting each title between the first keyword and the second keyword; according to the format of each title and the preset format corresponding to the directory at each level in each group of knowledge point hierarchical directories, the each title is divided into the directory of the corresponding level to obtain the multiple groups of knowledge point hierarchical directories.

[0133] In some embodiments, the text includes various text fragments, and the matching unit 702 matches the last-level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last-level directory under each group of knowledge point hierarchical directories, specifically including: performing the following steps for each group of knowledge point hierarchical directories: determining the similarity between the last-level directory under the group of knowledge point hierarchical directories and the various text fragments; determining the text fragment with the highest similarity to the last-level directory under the group of knowledge point hierarchical directories as the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories; or merging at least one text fragment whose similarity to the last-level directory under the group of knowledge point hierarchical directories is greater than or equal to a first similarity threshold to obtain the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories.

[0134] In some embodiments, after the text segment with the highest similarity to the last level directory under the group of knowledge point hierarchical directories is determined as the content corresponding to the last level directory under the group of knowledge point hierarchical directories, the matching unit 702 is also used to perform the following steps: if the similarity between the first text segment and the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories, then an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories is generated; wherein the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories is greater than or equal to a second similarity threshold.

[0135] The matching unit 702, after merging at least one text segment whose similarity with the last level directory under the group of knowledge point hierarchical directories is greater than or equal to the first similarity threshold to obtain the text content corresponding to the last level directory under the group of knowledge point hierarchical directories, the matching unit 702 is also used to perform the following steps: if the text segment corresponding to the similarity among the various similarities is less than the first similarity threshold and greater than or equal to the third similarity threshold, if the similarity between the second text segment and the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories, then an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories is generated; wherein the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories is less than the first similarity threshold and greater than or equal to the third similarity threshold.

[0136] The matching unit 702 contains key elements in each text segment, and the key elements include at least one of a title, a table, a picture, a paragraph and an annotation; wherein, when the matching unit 702 determines the similarity between the last level directory under the group of knowledge point hierarchical directories and the each text segment, it specifically includes: determining the similarity between the last level directory under the group of knowledge point hierarchical directories and each key element in each text segment; according to the weight coefficient corresponding to each key element, weighted summing up the similarities between the last level directory under the group of knowledge point hierarchical directories and each key element in each text segment, to obtain the similarity between the last level directory under the group of knowledge point hierarchical directories and each text segment.

[0137] In some embodiments, the teaching materials include a plurality of teaching materials, the teaching knowledge graph includes a respective teaching knowledge graph corresponding to each teaching material, and the construction unit 703 is further configured to perform the following steps: identifying entities representing the same object and relationships representing the same semantics in the plurality of teaching knowledge graphs; and associating the entities representing the same object and the relationships representing the same semantics in the plurality of teaching knowledge graphs to obtain a comprehensive teaching knowledge graph.

[0138] The knowledge graph construction apparatus provided in this embodiment belongs to the same application concept as the knowledge graph construction method provided in the above-mentioned embodiments of the present application, can perform the knowledge graph construction method provided in any of the above-mentioned embodiments of the present application, and has the corresponding functional modules and beneficial effects of performing the knowledge graph construction method. Technical details not described in detail in this embodiment can be found in the specific processing content of the knowledge graph construction method provided in the above-mentioned embodiments of the present application, which will not be described here again.

[0139] The functions implemented by the extraction unit 701, the matching unit 702, and the construction unit 703 described above can be implemented by the same or different processors, and the embodiments of the present application are not limited in this regard.

[0140] It should be understood that the units in the above apparatus can be implemented in the form of processor calling software. For example, the apparatus includes a processor connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units of the apparatus, where the processor can be a general processor such as a CPU or a microprocessor, and the memory can be an internal memory of the apparatus or an external memory of the apparatus. Alternatively, the units in the apparatus can be implemented in the form of hardware circuit. The functions of some or all of the units can be implemented by designing the hardware circuit. The hardware circuit can be understood as one or more processors. For example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all of the units are implemented by designing the logical relationship of elements in the circuit. For another example, in another implementation, the hardware circuit can be implemented by a PLD. Taking an FPGA as an example, it can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to implement the functions of some or all of the units. All the units of the above apparatus can be implemented in the form of processor calling software, or all the units can be implemented in the form of hardware circuit, or part of the units can be implemented in the form of processor calling software, and the remaining part can be implemented in the form of hardware circuit.

[0141] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and execute instructions, such as a CPU, a microprocessor, a GPU, or a DSP. In another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit may be fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the configuration of the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.

[0142] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0143] In addition, the various units in the above apparatus may be fully or partially integrated together, or may be implemented independently. In one implementation, these units are integrated together and implemented in the form of a system-on-chip (SOC). The SOC may include at least one processor for implementing any of the above methods or implementing the functions of the various units of the apparatus. The at least one processor may be of different types, such as a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0144] Exemplary electronic devices

[0145] The present application embodiment provides an electronic device, see Figure 8 As shown, the electronic device includes:

[0146] Memory 200 and processor 210;

[0147] The memory 200 is connected to the processor 210 and is used to store programs;

[0148] The processor 210 is used to implement the knowledge graph construction method disclosed in any of the above embodiments by running the program stored in the memory 200.

[0149] Specifically, the electronic device may further include: a bus, a communication interface 220 , an input device 230 and an output device 240 .

[0150] The processor 210, the memory 200, the communication interface 220, the input device 230 and the output device 240 are interconnected via a bus.

[0151] A bus may include a pathway that transfers information between components of a computer system.

[0152] Processor 210 can be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, or the like. It can also be an application-specific integrated circuit (ASIC), or one or more integrated circuits used to control the execution of the program of the present invention. It can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, or discrete hardware components.

[0153] The processor 210 may include a main processor, and may also include a baseband chip, a modem, and the like.

[0154] Memory 200 stores programs that implement the technical solutions of the present invention and may also store an operating system and other key services. Specifically, the programs may include program code, which includes computer operating instructions. More specifically, memory 200 may include read-only memory (ROM), other types of static storage devices capable of storing static information and instructions, random access memory (RAM), other types of dynamic storage devices capable of storing information and instructions, disk storage, flash memory, and the like.

[0155] The input device 230 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0156] Output device 240 may include devices that allow information to be output to a user, such as a display screen, printer, speakers, etc.

[0157] The communication interface 220 may include any device such as a transceiver to communicate with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0158] The processor 210 executes the program stored in the memory 200 and calls other devices, which can be used to implement the various steps of any knowledge graph construction method provided in the above embodiments of this application.

[0159] An embodiment of the present application also proposes a chip, which includes a processor and a data interface. The processor reads and runs the program stored in the memory through the data interface to execute the knowledge graph construction method introduced in any of the above embodiments. The specific processing process and its beneficial effects can be found in the embodiment introduction of the above-mentioned knowledge graph construction method.

[0160] Exemplary computer program products and storage media

[0161] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the knowledge graph construction method according to various embodiments of the present application described in any of the above embodiments of this specification.

[0162] The computer program product may be written in any combination of one or more programming languages ​​to implement the program code for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0163] In addition, an embodiment of the present application may also be a storage medium on which a computer program is stored. The computer program is executed by a processor to execute the steps of the knowledge graph construction method according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically can implement the steps of the above method embodiments.

[0164] For the sake of simplicity, the aforementioned method embodiments are described as a series of action combinations. However, those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in this specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0165] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similarities between the various embodiments can be referred to in conjunction with each other. For device embodiments, since they are generally similar to method embodiments, their description is relatively simple, and for relevant details, reference can be made to the description of the method embodiments.

[0166] The steps in the methods of each embodiment of the present application can be adjusted in sequence, merged, and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.

[0167] The modules and sub-modules in the devices and terminals of the various embodiments of the present application can be merged, divided, and deleted according to actual needs.

[0168] In the several embodiments provided in this application, it should be understood that the disclosed terminals, devices, and methods can be implemented in other ways. For example, the terminal embodiments described above are merely illustrative. For example, the division of modules or submodules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or module, which can be electrical, mechanical or other forms.

[0169] The modules or submodules described as separate components may or may not be physically separate, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place or distributed across multiple network modules or submodules. Some or all of the modules or submodules may be selected to achieve the purpose of this embodiment according to actual needs.

[0170] In addition, each functional module or submodule in each embodiment of the present application may be integrated into a processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into a single module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or software functional modules or submodules.

[0171] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0172] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, software executed by a processor, or a combination of the two. The software may be stored in random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0173] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0174] The above description of the disclosed embodiments will enable those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is to be construed in the widest manner consistent with the principles and novel features disclosed herein.

Claims

1. A knowledge graph construction method, characterized in that: include: Extracting the catalogue from the teaching materials to obtain multiple groups of knowledge point hierarchical catalogues, wherein each group of knowledge point hierarchical catalogues includes at least one first-level catalogue; Matching the last level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last level directory under each group of knowledge point hierarchical directories, wherein the text includes various text segments; The last level directory under each group of knowledge point hierarchical directories is matched with the text in the teaching materials to obtain the text content corresponding to the last level directory under each group of knowledge point hierarchical directories, including: Perform the following steps for each set of knowledge point classification catalogs: Determine the similarity between the last level directory under the group of knowledge point hierarchical directories and each of the text segments; If the similarity between the first text segment and the last level directory under another group of knowledge point hierarchical directories is greater than the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories, then generating an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories; wherein the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories is greater than or equal to a second similarity threshold; Based on the extraction of entities and entity relationships from the text content corresponding to the last level directory under each group of knowledge point hierarchical directories, a teaching knowledge graph is constructed.

2. The method according to claim 1, characterized in that The directory in the teaching materials is extracted to obtain multiple groups of knowledge point hierarchical directories, including: Identifying a first keyword and a second keyword in the teaching material, wherein the first keyword represents a starting position of the directory and the second keyword represents an ending position of the directory; extracting each title between the first keyword and the second keyword; According to the formats of the respective titles and the preset formats corresponding to the directories at each level in each group of knowledge point hierarchical directories, the respective titles are divided into directories at corresponding levels to obtain the plurality of groups of knowledge point hierarchical directories.

3. The method according to claim 1, characterized in that The step of matching the last level directory under each group of knowledge point hierarchical directories with the text in the teaching materials to obtain the text content corresponding to the last level directory under each group of knowledge point hierarchical directories further includes: Perform the following steps for each set of knowledge point classification catalogs: The text segment with the highest similarity to the last-level directory under the group of knowledge point hierarchical directories is determined as the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories; or, at least one text segment whose similarity to the last-level directory under the group of knowledge point hierarchical directories is greater than or equal to the first similarity threshold is merged to obtain the text content corresponding to the last-level directory under the group of knowledge point hierarchical directories.

4. The method according to claim 3, characterized in that After merging at least one text segment having a similarity greater than or equal to a first similarity threshold with the last level directory under the group of knowledge point hierarchical directories to obtain text content corresponding to the last level directory under the group of knowledge point hierarchical directories, the method further includes: If, among the respective similarities, the text segments corresponding to the similarities that are less than the first similarity threshold and greater than or equal to the third similarity threshold, and if the similarity between the second text segment and the last level directory under the other group of knowledge point hierarchical directories is greater than the similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories, then an association relationship is generated between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories; The similarity between the second text segment and the last level directory under the group of knowledge point hierarchical directories is less than a first similarity threshold and greater than or equal to a third similarity threshold.

5. The method according to claim 3 or 4, characterized in that Each text segment contains key elements, including at least one of a title, a table, a picture, a paragraph, and a note; The step of determining the similarity between the last level directory under the group of knowledge point hierarchical directories and each of the text segments includes: Determine the similarity between the last level directory under the knowledge point hierarchy directory and each key element in each text segment; According to the weight coefficients corresponding to the key elements, the similarities between the last level directory under the group of knowledge point hierarchical directories and the key elements in each text segment are weighted and summed to obtain the similarities between the last level directory under the group of knowledge point hierarchical directories and each text segment.

6. The method according to any one of claims 1 to 4, characterized in that The teaching materials include a plurality of teaching materials, the teaching knowledge graph includes a teaching knowledge graph corresponding to each teaching material, and the method further includes: Identify entities representing the same object and relationships representing the same semantics in multiple teaching knowledge graphs; The entities representing the same object in the multiple teaching knowledge graphs are associated, and the relationships representing the same semantics are associated to obtain a comprehensive teaching knowledge graph.

7. A knowledge graph construction device, characterized in that: include: An extraction unit is used to extract the catalogue from the teaching materials to obtain multiple groups of knowledge point hierarchical catalogues, wherein each group of knowledge point hierarchical catalogues includes at least one first-level catalogue; a matching unit, configured to match the last level directory under each group of knowledge point hierarchical directories with the main text in the teaching material, and obtain the main text content corresponding to the last level directory under each group of knowledge point hierarchical directories, wherein the main text includes various text segments; The matching unit matches the last level directory under each group of knowledge point hierarchical directories with the text in the teaching material to obtain the text content corresponding to the last level directory under each group of knowledge point hierarchical directories, including: Perform the following steps for each set of knowledge point classification catalogs: Determine the similarity between the last level directory under the group of knowledge point hierarchical directories and each of the text segments; If the similarity between the first text segment and the last level directory under another group of knowledge point hierarchical directories is greater than the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories, then generating an association relationship between the other group of knowledge point hierarchical directories and the group of knowledge point hierarchical directories; wherein the similarity between the first text segment and the last level directory under the group of knowledge point hierarchical directories is greater than or equal to a second similarity threshold; The construction unit is used to extract entities and entity relationships based on the text content corresponding to the last level directory under each group of knowledge point hierarchical directories to construct a teaching knowledge graph.

8. An electronic device, characterized in that: including memory and processor; The memory is connected to the processor and is used to store programs; The processor is configured to implement the method according to any one of claims 1 to 7 by running the program in the memory.

9. A computer program product, characterized in that The method comprises computer program instructions, which, when executed by a processor, cause the processor to implement the method according to any one of claims 1 to 7.

Images