A course knowledge graph automatic construction system based on large models

Abstract

The present invention discloses a system for automatically constructing a course knowledge graph based on a large model, which belongs to the technical field of paper evaluation systems. The system includes a login processing unit, a paper upload evaluation and ranking system, multiple school paper knowledge graph storage systems, a retrieval and access index system, and a graph creation unit. The login processing unit is in communication with the paper upload evaluation and ranking system and multiple school paper knowledge graph storage systems. The paper upload evaluation system is in communication with the school paper knowledge graph storage system. The retrieval and access index system is in communication with the school paper knowledge graph storage system and the graph creation unit. Each school paper knowledge graph storage system corresponds to a school, and the school paper knowledge graph storage system stores a course catalog. Through the above-mentioned method, the technology of the present invention realizes the creation of a course knowledge graph in combination with students' course papers, and at the same time realizes academic exchanges of course papers between different schools.

Description

Technical Field

[0001] The present invention relates to the technical field of paper evaluation systems, and in particular to a system for automatically constructing course knowledge graphs based on a large model. Background Art

[0002] The course knowledge map is the school's design of a student learning plan based on the course catalog, which makes the relationship between course design and learning clearer and more visible.

[0003] Among many existing technologies, Chinese patent application CN114911949A discloses a method and system for constructing a course knowledge graph, which includes generating an outline knowledge graph based on a teaching outline and a preset weight value; generating a lesson plan knowledge graph based on a lesson plan; establishing a vector representation between knowledge points with the core knowledge points in the outline knowledge graph as the root node; using the cosine similarity formula to calculate the similarity between the vectors of entities in the outline knowledge graph and the lesson plan knowledge graph; if the similarity is greater than or equal to a set threshold, it is considered to be the same entity, and the course knowledge points are determined based on the knowledge points in the lesson plan knowledge graph, and the subgraph with the knowledge points in the lesson plan knowledge graph as the root node is stored in the corresponding position of the course knowledge graph; if the similarity is less than the set threshold, it is considered to be different entities, and the subgraphs with the two entities as root nodes are respectively stored in the corresponding positions of the course knowledge graph; and integrated to form a complete course knowledge graph.

[0004] However, this patent application cannot create a course knowledge graph by combining students' course papers, nor can it achieve academic exchanges of course papers between different schools.

[0005] Based on this, the present invention designs a large-model-based automatic construction system for course knowledge graphs to solve the above problems. Summary of the Invention

[0006] In response to the above-mentioned shortcomings of the existing technology, the present invention provides an automatic construction system of course knowledge graph based on a large model.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] A large-scale model-based course knowledge graph automatic construction system, including a login processing unit, a paper upload evaluation and ranking system, a multi-school paper knowledge graph storage system, a retrieval and indexing system, and a graph creation unit;

[0009] The login processing unit is connected to the paper upload evaluation and ranking system and the paper knowledge graph storage system of multiple schools;

[0010] The paper upload evaluation system is connected to the school paper knowledge graph storage system;

[0011] The retrieval and indexing system is connected to the school's paper knowledge graph storage system and graph creation unit;

[0012] Each school paper knowledge graph storage system corresponds to a school, and the school paper knowledge graph storage system stores the course catalog;

[0013] The login processing unit is used for student user's identity information verification login;

[0014] The identity information corresponds to at least one school;

[0015] The paper upload evaluation and ranking system is used for student users who have completed identity authentication to upload course papers to the login processing unit, which then sends the course papers to the corresponding school paper knowledge graph storage system for storage;

[0016] The paper upload evaluation and ranking system is also used to score papers and generate school rankings and overall rankings based on paper scores, course catalogs, and paper keywords;

[0017] The graph creation unit is used to create a course paper knowledge graph based on the course catalog, school ranking table and overall ranking table;

[0018] The retrieval and indexing system is used to retrieve keywords. It is also used to match the retrieval keywords with the course catalog and, when the pairing is successful, to create an index path 1 between the retrieval keywords and the course paper knowledge graph entity.

[0019] The retrieval and indexing system is also used to create an index path between the course paper knowledge graph and the school paper knowledge graph storage system to store the papers.

[0020] The retrieval and indexing system accesses the knowledge graph of the school's course papers and the paper texts through index path one and index path two respectively;

[0021] It also includes a paper sharing server, which is connected to the login processing unit, multiple school paper knowledge graph storage systems, and a retrieval index system;

[0022] The paper sharing server is used to share paper data with multiple school paper knowledge graph storage systems;

[0023] The retrieval index system is also used to create an index path between the course paper knowledge graph and the papers shared by the paper sharing server;

[0024] The retrieval and indexing system accesses the paper texts of all schools through index path three.

[0025] Furthermore, the login processing unit includes a central processing server and a user verification login module;

[0026] The central processing server is connected to the user verification login module, multiple school paper knowledge graph storage systems, and paper upload evaluation and ranking systems;

[0027] The user verification login module is used for student users to verify their student ID number and log in;

[0028] The central processing server is used to receive course papers sent by the paper upload evaluation and ranking system and send the course papers to the corresponding school paper knowledge graph storage system.

[0029] Furthermore, the paper upload evaluation and ranking system includes a paper upload module, a paper evaluation system, and a paper ranking module;

[0030] The paper upload module is connected to the central processing server in communication, the paper evaluation system is connected to the central processing server and the paper ranking module in communication, and the paper ranking module is connected to the graph creation unit in communication;

[0031] The paper upload module is used for student users to upload course papers. The paper upload module sends the course papers to the central processing server, and the central processing server sends the course papers to the paper evaluation system and the corresponding school paper knowledge graph storage system;

[0032] The paper evaluation system is used to extract the keyword L of the paper based on the natural language model and match L with the keyword Z in the course catalog. The paper evaluation system sends the matched keyword L to the paper ranking module;

[0033] The paper evaluation system scores papers based on the paper evaluation model;

[0034] The paper ranking module is used to generate the overall ranking table and school ranking table:

[0035] The paper ranking module collects a collection of papers with keyword L from a school, ranks the paper titles from high to low according to the paper scores, and uses keyword L as the ranking theme O of the school ranking table. The paper ranking module sends the school ranking table to the graph creation unit;

[0036] The paper ranking module collects all the papers with keyword L from all schools, ranks the paper scores from high to low, and uses keyword L as the ranking theme P of the overall ranking table. The paper ranking module sends the overall ranking table to the graph creation unit.

[0037] Furthermore, the school paper knowledge graph storage system includes the school paper database and the school course server;

[0038] The school's paper database is connected to the central processing server, paper sharing server, and retrieval and indexing system; the school's course server is connected to the atlas creation unit and retrieval and indexing system;

[0039] The school paper database is used to receive and store course papers sent by the central processing server;

[0040] The school course server is used to store the course paper knowledge graph and course catalog. The school course server sends the course catalog to the graph creation unit and the retrieval index system.

[0041] The paper sharing server shares paper data with multiple school paper databases, and the central processing server is used for data sharing authorization management between school paper databases and the paper sharing server.

[0042] Furthermore, the graph creation unit includes a course paper knowledge graph creation module;

[0043] The course paper knowledge graph creation module is connected to the school course server and paper ranking module;

[0044] The course paper knowledge graph creation module is used to automatically construct the course paper knowledge graph and adopts the following steps:

[0045] Step A1: The course paper knowledge graph creation module reads the course catalog stored on the school course server and identifies the keyword Z in the catalog content. The course paper knowledge graph creation module identifies the relationship C between the keyword Z in the course catalog based on the natural language model. The course paper knowledge graph creation module uses the keyword Z as the entity z and forms a triplet 1 through the two entities z and the relationship C.

[0046] Step A2: The course paper knowledge graph creation module receives the school ranking list sent by the paper ranking module and identifies the ranking topic B in the school ranking list. The course paper knowledge graph creation module matches B with entity z based on the natural language model. The course paper knowledge graph creation module extracts the top three paper titles in the school ranking list and uses them as entity V. At the same time, it creates a relationship N between entity V and the corresponding entity z, and forms triple two through entity V, entity z, and relationship N.

[0047] Step A3: The course paper knowledge graph creation module uses the school ranking table and the overall ranking table as entity G1 and entity G2, respectively, and creates the relationship R1 between entity G1 and entity z, and the relationship R2 between entity G2 and entity z. The entity G1, entity z, and relationship R1 form triple three, and the entity G2, entity z, and relationship R2 form triple four.

[0048] Step A4, the course paper knowledge graph creation module creates a course paper knowledge graph by combining triple one, triple two, triple three, and triple four. The course paper knowledge graph creation module sends the course paper knowledge graph to the school course server.

[0049] Furthermore, the search and indexing system includes a keyword search module, a search module, an indexing module 1, an indexing module 2 and an indexing module 3;

[0050] The keyword search module is used by student users to search for keyword U, and the keyword search module sends the searched keyword U to the retrieval module;

[0051] The retrieval module matches keyword U with keyword Z in the course catalog;

[0052] When keyword U has a matching keyword Z, the review module sends keyword U to index module 1;

[0053] When keyword U does not have a matching keyword Z, the query module returns the invalid keyword U.

[0054] The indexing module 1 is used to create an index path 1 between the entity z of the course paper knowledge graph stored in the school course server and the keyword U, and the indexing module 1 sends the index path 1 to the retrieval module;

[0055] Index module two is used to create index path three between entity V of the course paper knowledge graph stored in the school course server and entity G1 (school ranking table) in the course paper knowledge graph, the paper title in entity G2 (total ranking table) and the paper text in multiple school paper databases called by the paper sharing server. Index module three sends index path three to the retrieval module; the retrieval module retrieves the location of the entity corresponding to the course paper knowledge graph through index path one.

[0056] Furthermore, the retrieval module retrieves the text of the school's papers through index path 2;

[0057] The retrieval module retrieves all school paper texts through index path three;

[0058] The retrieval module retrieves all school paper texts through index path four.

[0059] Furthermore, the graph creation unit also includes a paper retrieval knowledge graph creation module;

[0060] The paper retrieval knowledge graph creation module is connected to the retrieval module;

[0061] The paper retrieval knowledge graph creation module is used for automatic construction of the paper retrieval knowledge graph.

[0062] Furthermore, the retrieval module is also used to generate a retrieval record table of the subject papers corresponding to the keyword U, and the retrieval module sends the retrieval record table to the paper retrieval knowledge graph creation module;

[0063] The access record table includes the title of the paper and the time of access;

[0064] The paper retrieval knowledge graph creation module is connected to the personal database, and the personal database is connected to the retrieval module;

[0065] The paper retrieval knowledge graph creation module extracts paper titles based on the access record table. The course paper knowledge graph creation module sorts the paper titles along the timeline according to the access time, and uses the paper titles as entities F. At the same time, it creates a relationship W between the two entities F. The two entities F and the relationship W form a triple five. The paper retrieval knowledge graph creation module generates a paper retrieval knowledge graph based on the triple five. The paper retrieval knowledge graph creation module sends the paper retrieval knowledge graph to the personal database.

[0066] Personal database is used to store knowledge graph for paper retrieval;

[0067] The index module 3 is also used to create an index path 4 between the entity F stored in the paper retrieval knowledge graph in the index module 3 and the paper text in the multiple school paper databases called by the paper sharing server. The index module 3 sends the index path 4 to the retrieval module;

[0068] The retrieval module retrieves the paper retrieval knowledge graph in the personal database and retrieves the paper text corresponding to entity F through index path four.

[0069] Furthermore, it also includes a ranking update module, which is connected to the paper ranking module and the course paper knowledge graph creation module;

[0070] The ranking update module is used to send ranking update signals to the paper ranking module at intervals of one hour. After receiving the ranking update signal, the paper ranking module updates the school ranking table and the overall ranking table, and sends the updated school ranking table and overall ranking to the course paper knowledge graph creation module.

[0071] Compared with the existing technology, the present invention has the following advantages: 1. It connects multiple school thesis knowledge graph storage systems through the login processing unit, so that students from different schools can log in to the system and send the course papers uploaded by students to the corresponding school thesis knowledge graph storage system, thereby facilitating the school's management of the students' subject papers;

[0072] Through the paper upload evaluation and ranking system, the papers of this subject are scored based on the paper large model evaluation algorithm, and a school ranking table and an overall ranking table matching the course catalog are generated, so that it is easy to understand the ranking of course papers during course learning;

[0073] Through the paper sharing server, data sharing between the paper knowledge graph storage systems of different schools is achieved, thus facilitating academic exchanges between schools.

[0074] The graph creation unit automatically creates a course paper knowledge graph by integrating the ranking of the school's students' thesis papers, the ranking of all schools, and the school's course catalog. This allows the school's course catalog to be combined with the school's students' thesis papers and the ranking list of the thesis papers, and then displayed in the form of a knowledge graph.

[0075] Through the retrieval and access index system, students can quickly access the knowledge map of their school's course papers, as well as their school's project papers and other schools' project papers.

[0076] 2. When the present invention is used, the paper data is shared between the paper sharing server and multiple school paper databases, thereby creating an academic exchange channel between schools. At the same time, students are only supported to access the course paper knowledge graph of their own school, thereby ensuring that the school's course schedule is kept confidential.

[0077] 3. When the present invention is used, index path one, index path two, index path three and index path four created by index module one, index module two and index module three realize efficient access to data for student users. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] To more clearly illustrate the embodiments of the present invention 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 only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0079] Figure 1 This is a block diagram of a system for automatically constructing a course knowledge graph based on a large model according to the present invention;

[0080] Figure 2 A partial schematic diagram of the course paper knowledge graph of the present invention;

[0081] Figure 3 A partial schematic diagram of the paper retrieval knowledge graph of the present invention;

[0082] Figure 4 It is a schematic diagram of the school ranking table and the overall ranking table of the present invention.

[0083] The numbers in the figure represent:

[0084] 1. Central processing server 2. Paper upload module 3. User verification and login module 4. Keyword search module 5. Paper evaluation system 6. Paper ranking module 7. Course paper knowledge graph creation module 8. School paper database 9. Paper sharing server 10. Retrieval module 11. School course server 12. Paper retrieval knowledge graph creation module 13. Personal database 14. Index module 1 15. Index module 2 16. Index module 3 17. Ranking update module DETAILED DESCRIPTION

[0085] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0086] Example 1: In some embodiments, please refer to the accompanying drawings of the specification. Figure 1-Figure 4 , a course knowledge graph automatic construction system based on a large model, including a login processing unit, a paper upload evaluation and ranking system, a multi-school paper knowledge graph storage system, a retrieval and indexing system, and a graph creation unit;

[0087] The login processing unit is connected to the paper upload evaluation and ranking system and the paper knowledge graph storage system of multiple schools;

[0088] The paper upload evaluation system is connected to the school paper knowledge graph storage system;

[0089] The retrieval and indexing system is connected to the school's paper knowledge graph storage system and graph creation unit;

[0090] Each school paper knowledge graph storage system corresponds to a school, and the school paper knowledge graph storage system stores the course catalog;

[0091] The login processing unit is used for student user's identity information verification login;

[0092] The identity information corresponds to at least one school;

[0093] The paper upload evaluation and ranking system is used for student users who have completed identity authentication to upload course papers to the login processing unit, which then sends the course papers to the corresponding school paper knowledge graph storage system for storage;

[0094] The paper upload evaluation and ranking system is also used to score papers and generate school rankings and overall rankings based on paper scores, course catalogs, and paper keywords;

[0095] The graph creation unit is used to create a course paper knowledge graph based on the course catalog, school ranking table and overall ranking table;

[0096] The retrieval and indexing system is used to retrieve keywords. It is also used to match the retrieval keywords with the course catalog and, when the pairing is successful, to create an index path 1 between the retrieval keywords and the course paper knowledge graph entity.

[0097] The retrieval and indexing system is also used to create an index path between the course paper knowledge graph and the school paper knowledge graph storage system to store the papers.

[0098] The retrieval and indexing system accesses the knowledge graph of the school's course papers and the paper texts through index path one and index path two respectively;

[0099] It also includes a paper sharing server 9, which is connected to the login processing unit, multiple school paper knowledge graph storage systems, and a retrieval index system;

[0100] The paper sharing server 9 is used to share paper data with multiple school paper knowledge graph storage systems;

[0101] The retrieval and indexing system is also used to create an index path three between the course paper knowledge graph and the papers shared by the paper sharing server 9;

[0102] The retrieval and indexing system accesses the paper texts of all schools through index path three.

[0103] When the present invention is used, the student user completes the verification and login of the identity information through the login processing unit;

[0104] Through the login processing unit communication, multiple school thesis knowledge graph storage systems are connected, so that students from different schools can log in to the system and send the course papers uploaded by students to the corresponding school thesis knowledge graph storage system, so as to facilitate the school to manage the subject papers of its students;

[0105] Through the paper upload evaluation and ranking system, the papers of this subject are scored based on the paper large model evaluation algorithm, and a school ranking table and an overall ranking table matching the course catalog are generated, so that it is easy to understand the ranking of course papers during course learning;

[0106] Through the paper sharing server 9, data sharing between the paper knowledge graph storage systems of different schools is achieved, thereby facilitating academic exchanges between schools.

[0107] The graph creation unit automatically creates a course paper knowledge graph by integrating the ranking of the school's students' thesis papers, the ranking of all schools, and the school's course catalog. This allows the school's course catalog to be combined with the school's students' thesis papers and the ranking list of the thesis papers, and then displayed in the form of a knowledge graph.

[0108] Through the retrieval and access index system, students can quickly access the knowledge map of their school's course papers, as well as their school's project papers and other schools' project papers.

[0109] Embodiment 2: In some embodiments, as Figure 1-Figure 4 As shown, as a preferred embodiment of the present invention, the login processing unit includes a central processing server 1 and a user verification login module 3;

[0110] The central processing server 1 is connected to the user verification login module 3, multiple school paper knowledge graph storage systems, and paper upload evaluation and ranking systems;

[0111] User verification login module 3 is used for student user's student ID verification login;

[0112] The central processing server 1 is used to receive the course papers sent by the paper upload evaluation and ranking system and send the course papers to the corresponding school paper knowledge graph storage system;

[0113] The paper upload evaluation and ranking system includes a paper upload module 2, a paper evaluation system 5, and a paper ranking module 6;

[0114] The paper uploading module 2 is in communication with the central processing server 1, the paper evaluation system 5 is in communication with the central processing server 1 and the paper ranking module 6, and the paper ranking module 6 is in communication with the atlas creation unit;

[0115] The paper upload module 2 is used for student users to upload course papers. The paper upload module 2 sends the course papers to the central processing server 1. The central processing server 1 sends the course papers to the paper evaluation system 5 and the corresponding school paper knowledge graph storage system;

[0116] The paper evaluation system 5 is used to extract the keyword L of the paper based on the natural language model and match L with the keyword Z of the course catalog. The paper evaluation system 5 sends the matched keyword L to the paper ranking module 6;

[0117] The paper evaluation system 5 scores the paper based on the paper evaluation model;

[0118] In this invention application, the paper evaluation model adopts the existing mature AI graduation thesis quality evaluation system;

[0119] Paper Ranking Module 6 is used to generate the overall ranking table and school ranking table:

[0120] The paper ranking module 6 collects a collection of papers with keyword L from a school, ranks the paper titles from high to low according to the paper scores, and uses keyword L as the ranking theme O of the school ranking table. The paper ranking module 6 sends the school ranking table to the graph creation unit;

[0121] The paper ranking module 6 collects all papers with keyword L from all schools, ranks the papers by scores from high to low, and uses keyword L as the ranking theme P of the overall ranking table. The paper ranking module 6 sends the overall ranking table to the graph creation unit;

[0122] The school paper knowledge graph storage system includes a school paper database 8 and a school course server 11;

[0123] The school paper database 8 is connected to the central processing server 1, the paper sharing server 9, and the search and indexing system. The school course server 11 is connected to the atlas creation unit and the search and indexing system.

[0124] The school paper database 8 is used to receive and store course papers sent by the central processing server 1;

[0125] The school course server 11 is used to store the course paper knowledge graph and course catalog. The school course server 11 sends the course catalog to the graph creation unit and the retrieval index system.

[0126] The paper sharing server 9 shares paper data with multiple school paper databases 8, and the central processing server 1 is used for data sharing authorization management between the school paper database 8 and the paper sharing server 9;

[0127] When the present invention is used, the paper data is shared between the paper sharing server 9 and multiple school paper databases 8, thereby creating an academic exchange channel between schools. At the same time, students are only supported to access the course paper knowledge graph of their own school, thereby ensuring that the school course schedule is kept confidential.

[0128] The graph creation unit includes a course paper knowledge graph creation module 7;

[0129] The course paper knowledge graph creation module 7 is connected to the school course server 11 and the paper ranking module 6;

[0130] The course paper knowledge graph creation module 7 is used to automatically construct the course paper knowledge graph and adopts the following steps:

[0131] Step A1: The course paper knowledge graph creation module 7 reads the course catalog stored in the school course server 11 and identifies the keyword Z in the catalog content. The course paper knowledge graph creation module 7 identifies the relationship C between the keyword Z in the course catalog based on the natural language model. The course paper knowledge graph creation module 7 uses the keyword Z as the entity z and forms a triplet 1 through the two entities z and the relationship C.

[0132] Step A2: The course paper knowledge graph creation module 7 receives the school ranking list sent by the paper ranking module 6 and identifies the ranking topic B in the school ranking list. The course paper knowledge graph creation module 7 matches B with entity z based on the natural language large model. The course paper knowledge graph creation module 7 extracts the top three paper titles in the school ranking list and uses them as entity V. At the same time, it creates a relationship N between entity V and the corresponding entity z, and forms triple two through entity V, entity z, and relationship N.

[0133] Step A3: The course paper knowledge graph creation module 7 takes the school ranking table and the overall ranking table as entity G1 and entity G2, respectively, and creates a relationship R1 between entity G1 and entity z, and a relationship R2 between entity G2 and entity z. The entity G1, entity z, and relationship R1 form a triplet three, and the entity G2, entity z, and relationship R2 form a triplet four.

[0134] Step A4, the course paper knowledge graph creation module 7 creates a course paper knowledge graph by combining triple one, triple two, triple three, and triple four, and the course paper knowledge graph creation module 7 sends the course paper knowledge graph to the school course server 11.

[0135] The search and indexing system includes a keyword search module 4, a search module 10, an indexing module 14, an indexing module 2 15 and an indexing module 3 16;

[0136] The keyword search module 4 is used for student users to search for keywords U, and the keyword search module 4 sends the searched keywords U to the review module 10;

[0137] The review module 10 matches the keyword U with the keyword Z in the course catalog;

[0138] When keyword U has a matching keyword Z, the review module 10 sends keyword U to the index module 14;

[0139] When there is no matching keyword Z for keyword U, the review module 10 returns the feedback that keyword U is invalid;

[0140] The indexing module 14 is used to create an index path 1 between the entity z of the course paper knowledge graph stored in the school course server 11 and the keyword U, and the indexing module 14 sends the index path 1 to the retrieval module 10;

[0141] The second indexing module 15 is used to create a second index path between the entity V of the course paper knowledge graph stored in the school course server 11 and the paper title in the entity G1 (school ranking table) and the paper text stored in the school paper database 8. The second indexing module 15 sends the second index path to the retrieval module 10;

[0142] The indexing module 3 16 is used to create an index path 3 between the paper title in the entity G2 (total ranking table) of the course paper knowledge graph stored in the school course server 11 and the paper text in the multiple school paper databases 8 called by the paper sharing server 9, and the indexing module 3 16 sends the index path 3 to the retrieval module 10;

[0143] The retrieval module 10 retrieves the location of the entity corresponding to the course paper knowledge graph through the index path 1;

[0144] The access module 10 accesses the school's paper text through the index path 2;

[0145] The access module 10 accesses all school paper texts through the index path 3;

[0146] The access module 10 accesses all school paper texts through index path 4.

[0147] The graph creation unit also includes a paper retrieval knowledge graph creation module 12;

[0148] The paper retrieval knowledge graph creation module 12 is in communication with the retrieval module 10;

[0149] The paper retrieval knowledge graph creation module 12 is used for automatically constructing the paper retrieval knowledge graph.

[0150] The review module 10 is also used to generate a review record table for the subject papers corresponding to the keyword U. The review module 10 sends the review record table to the paper retrieval knowledge graph creation module 12.

[0151] The access record table includes the title of the paper and the time of access.

[0152] The paper retrieval knowledge graph creation module 12 is in communication with the personal database 13, and the personal database 13 is in communication with the retrieval module 10;

[0153] The paper retrieval knowledge graph creation module 12 extracts the paper titles according to the access record table. The course paper knowledge graph creation module 7 sorts the paper titles along the timeline according to the access time, and uses the paper titles as entities F. At the same time, a relationship W between two entities F is created. The two entities F and the relationship W form a triple five. The paper retrieval knowledge graph creation module 12 generates a paper retrieval knowledge graph based on the triple five. The paper retrieval knowledge graph creation module 12 sends the paper retrieval knowledge graph to the personal database 13.

[0154] The personal database 13 is used to store the paper retrieval knowledge graph;

[0155] The index module 3 16 is further configured to create an index path 4 between the entity F stored in the paper retrieval knowledge graph in the index module 3 16 and the paper texts in the multiple school paper databases 8 called by the paper sharing server 9, and the index module 3 16 sends the index path 4 to the retrieval module 10;

[0156] The access module 10 accesses the paper retrieval knowledge graph in the personal database 13, and accesses the paper text corresponding to the entity F through the index path four.

[0157] When the present invention is used, index path one, index path two, index path three and index path four created by index module one 14, index module two 15 and index module three 16 realize efficient access to data for student users.

[0158] When the present invention is used, the paper retrieval knowledge map creation module 12 is used to create a paper retrieval knowledge map based on the student's personal subject paper reading record, thereby facilitating students to clarify their own learning path in course learning and improving course learning efficiency;

[0159] Embodiment 3: In some embodiments, as Figure 1-Figure 4 As shown, as a preferred embodiment of the present invention, it also includes a ranking update module 17, and the ranking update module 17 is communicatively connected with the paper ranking module 6 and the course paper knowledge graph creation module 7;

[0160] The ranking update module 17 is used to send a ranking update signal to the paper ranking module 6 at intervals of one hour. After receiving the ranking update signal, the paper ranking module 6 updates the school ranking table and the overall ranking table, and sends the updated school ranking table and overall ranking to the course paper knowledge graph creation module 7.

[0161] When the present invention is used, real-time updating of ranking data is achieved through the ranking updating module 17 .

[0162] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A system for automatically constructing a course knowledge graph based on a large model, characterized by: It includes a login processing unit, a paper upload evaluation and ranking system, a multi-school paper knowledge graph storage system, a retrieval index system, and a graph creation unit; The login processing unit is connected to the paper upload evaluation and ranking system and the paper knowledge graph storage system of multiple schools; The paper upload evaluation system is connected to the school paper knowledge graph storage system; The retrieval and indexing system is connected to the school's paper knowledge graph storage system and graph creation unit; Each school paper knowledge graph storage system corresponds to a school, and the school paper knowledge graph storage system stores the course catalog; The login processing unit is used for student user's identity information verification login; The identity information corresponds to at least one school; The paper upload evaluation and ranking system is used for student users who have completed identity authentication to upload course papers to the login processing unit, which then sends the course papers to the corresponding school paper knowledge graph storage system for storage; The paper upload evaluation and ranking system is also used to score papers and generate school rankings and overall rankings based on paper scores, course catalogs, and paper keywords; The graph creation unit is used to create a course paper knowledge graph based on the course catalog, the school ranking table and the overall ranking table; the graph creation unit includes a course paper knowledge graph creation module (7); the paper upload evaluation ranking system includes a paper ranking module (6), and the school paper knowledge graph storage system includes a school course server (11); the course paper knowledge graph creation module (7) is connected to the school course server (11) and the paper ranking module (6); the course paper knowledge graph creation module (7) is used for automatically constructing the course paper knowledge graph, and adopts the following steps: Step A1, the course paper knowledge graph creation module (7) reads the course catalog stored in the school course server (11) and identifies the keyword Z in the catalog content. The course paper knowledge graph creation module (7) identifies the relationship C between the keyword Z in the course catalog based on the natural language big model. The course paper knowledge graph creation module (7) takes the keyword Z as the entity z and forms a triplet 1 through the two entities z and the relationship C; Step A2, the course paper knowledge graph creation module (7) receives the school ranking list sent by the paper ranking module (6), and identifies the ranking topic B in the school ranking list. The course paper knowledge graph creation module (7) matches B with the entity z based on the natural language large model. The course paper knowledge graph creation module (7) extracts the top three paper titles in the school ranking list and uses them as entity V. At the same time, it creates a relationship N between entity V and the corresponding entity z, and forms a triple two through entity V, entity z, and relationship N. Step A3, the course paper knowledge graph creation module (7) takes the school ranking table and the total ranking table as entity G1 and entity G2 respectively, and creates the relationship R1 between entity G1 and entity z and the relationship R2 between entity G2 and entity z, and forms triple three through entity G1, entity z, and relationship R1, and triple four through entity G2, entity z, and relationship R2; Step A4, the course paper knowledge graph creation module (7) creates a course paper knowledge graph by combining triple 1, triple 2, triple 3, and triple 4, and the course paper knowledge graph creation module (7) sends the course paper knowledge graph to the school course server (11); The retrieval and indexing system is used to retrieve keywords. It is also used to match the retrieval keywords with the course catalog and, when the pairing is successful, to create an index path 1 between the retrieval keywords and the course paper knowledge graph entity. The retrieval and indexing system is also used to create an index path between the course paper knowledge graph and the school paper knowledge graph storage system to store the papers. The retrieval and indexing system accesses the knowledge graph of the school's course papers and the paper texts through index path one and index path two respectively; It also includes a paper sharing server (9), which is connected to the login processing unit, multiple school paper knowledge graph storage systems, and a retrieval index system; The paper sharing server (9) is used to share paper data with multiple school paper knowledge graph storage systems; The retrieval and indexing system is also used to create an index path between the course paper knowledge graph and the papers shared by the paper sharing server (9); The retrieval and indexing system accesses the paper texts of all schools through index path three.

2. The system for automatically constructing a course knowledge graph based on a large model according to claim 1 is characterized in that: The login processing unit includes a central processing server (1) and a user verification login module (3); The central processing server (1) is connected to the user verification login module (3), multiple school paper knowledge graph storage systems, and the paper upload evaluation and ranking system; User verification login module (3) is used for student user's student ID verification login; The central processing server (1) is used to receive the course papers sent by the paper upload evaluation and ranking system and send the course papers to the corresponding school paper knowledge graph storage system.

3. The automatic construction system of course knowledge graph based on large model according to claim 2 is characterized in that: The paper upload evaluation and ranking system also includes a paper upload module (2) and a paper evaluation system (5); The paper uploading module (2) is connected to the central processing server (1), the paper evaluation system (5) is connected to the central processing server (1) and the paper ranking module (6), and the paper ranking module (6) is connected to the graph creation unit; The paper upload module (2) is used for student users to upload course papers. The paper upload module (2) sends the course papers to the central processing server (1). The central processing server (1) sends the course papers to the paper evaluation system (5) and the corresponding school paper knowledge graph storage system; The paper evaluation system (5) is used to extract the keyword L of the paper based on the natural language large model and match L with the keyword Z of the course catalog. The paper evaluation system (5) sends the matched keyword L to the paper ranking module (6); The paper evaluation system (5) scores the papers based on the paper evaluation model; The paper ranking module (6) is used to generate the overall ranking table and the school ranking table: The paper ranking module (6) collects the papers of a school with the keyword L and ranks the paper titles from high to low according to the paper scores, and uses the keyword L as the ranking theme O of the school ranking table. The paper ranking module (6) sends the school ranking table to the graph creation unit; The paper ranking module (6) collects all the papers with keyword L from all schools, ranks the paper scores from high to low, and uses keyword L as the ranking theme P of the overall ranking table. The paper ranking module (6) sends the overall ranking table to the graph creation unit.

4. The system for automatically constructing a course knowledge graph based on a large model according to claim 3 is characterized in that: The school paper knowledge graph storage system also includes the school paper database (8); The school paper database (8) is connected to the central processing server (1), the paper sharing server (9), and the search and access index system, and the school course server (11) is connected to the atlas creation unit and the search and access index system; The school paper database (8) is used to receive and store course papers sent by the central processing server (1); The school course server (11) is used for storing the course paper knowledge graph and the course catalog. The school course server (11) sends the course catalog to the graph creation unit and the retrieval index system; The paper sharing server (9) shares paper data with multiple school paper databases (8), and the central processing server (1) is used for data sharing authorization management between the school paper database (8) and the paper sharing server (9).

5. The system for automatically constructing a course knowledge graph based on a large model according to claim 4 is characterized in that: The search and indexing system includes a keyword search module (4), a search module (10), an indexing module 1 (14), an indexing module 2 (15) and an indexing module 3 (16); The keyword search module (4) is used for student users to search for keywords U, and the keyword search module (4) sends the searched keywords U to the review module (10); The review module (10) matches the keyword U with the keyword Z in the course catalog; When keyword U has a matching keyword Z, the review module (10) sends keyword U to the index module 1 (14); When there is no matching keyword Z for keyword U, the review module (10) returns a feedback that the keyword U is invalid; The indexing module 1 (14) is used to create an index path 1 between the entity z of the course paper knowledge graph stored in the school course server (11) and the keyword U, and the indexing module 1 (14) sends the index path 1 to the retrieval module (10); The index module 2 (15) is used to create an index path 2 between the entity V of the course paper knowledge graph stored in the school course server (11) and the paper title in the entity G1 (school ranking table) and the paper text stored in the school paper database (8), and the index module 2 (15) sends the index path 2 to the retrieval module (10); The index module three (16) is used to create an index path three between the paper title in the entity G2 (total ranking table) of the course paper knowledge graph stored in the school course server (11) and the paper text in the multiple school paper databases (8) called by the paper sharing server (9). The index module three (16) sends the index path three to the retrieval module (10); the retrieval module (10) retrieves the location of the entity corresponding to the course paper knowledge graph through the index path one.

6. The system for automatically constructing a course knowledge graph based on a large model according to claim 5 is characterized in that: The access module (10) accesses the school's paper texts through index path 2; The retrieval module (10) retrieves all school paper texts through index path three; The access module (10) accesses all school paper texts through index path four.

7. The system for automatically constructing a course knowledge graph based on a large model according to claim 6 is characterized in that: The graph creation unit also includes a paper retrieval knowledge graph creation module (12); The paper retrieval knowledge graph creation module (12) is connected to the retrieval module (10); The paper retrieval knowledge graph creation module (12) is used to automatically construct the paper retrieval knowledge graph.

8. The system for automatically constructing a course knowledge graph based on a large model according to claim 7 is characterized in that: The retrieval module (10) is further used to generate a retrieval record table of the subject papers corresponding to the keyword U, and the retrieval module (10) sends the retrieval record table to the paper retrieval knowledge graph creation module (12); The access record table includes the title of the paper and the time of access; The paper retrieval knowledge graph creation module (12) is connected to the personal database (13), and the personal database (13) is connected to the retrieval module (10); The paper retrieval knowledge graph creation module (12) extracts the paper titles according to the access record table. The course paper knowledge graph creation module (7) sorts the paper titles along the timeline according to the access time, and takes the paper titles as entities F. At the same time, a relationship W between two entities F is created. A triple five is formed by the two entities F and the relationship W. The paper retrieval knowledge graph creation module (12) generates a paper retrieval knowledge graph based on the triple five. The paper retrieval knowledge graph creation module (12) sends the paper retrieval knowledge graph to the personal database (13). Personal database (13) is used to store the knowledge graph for paper retrieval; The index module three (16) is further configured to create an index path four between the entity F stored in the paper retrieval knowledge graph in the index module three (16) and the paper texts in the multiple school paper databases (8) called by the paper sharing server (9), and the index module three (16) sends the index path four to the retrieval module (10); The access module (10) accesses the paper retrieval knowledge graph in the personal database (13), and accesses the paper text corresponding to the entity F through the index path four.

9. The system for automatically constructing a course knowledge graph based on a large model according to claim 8 is characterized in that: It also includes a ranking update module (17), which is connected to the paper ranking module (6) and the course paper knowledge graph creation module (7); The ranking update module (17) is used to send a ranking update signal to the paper ranking module (6) at intervals of one hour. After receiving the ranking update signal, the paper ranking module (6) updates the school ranking table and the overall ranking table, and sends the updated school ranking table and overall ranking to the course paper knowledge graph creation module (7).

Images