Semantic network-based document relationship analysis apparatus and method

Abstract

The present invention relates to a semantic network-based document relationship analysis apparatus that analyzes a relationship between documents by utilizing artificial intelligence and natural language processing so as to identify an association between the documents. The semantic network-based document relationship analysis apparatus comprises: a document data reception unit that receives a plurality of document data from a user terminal; a document data analysis unit that analyzes text content of each document data by using a natural language processing technique and a text mining method, and extracts main concepts and keywords of the document data; a semantic network generation unit that analyzes the main concepts and keywords extracted from each document data, and generates a semantic network between the corresponding document data when the main concepts and keywords are determined to have a synonym relationship, a hierarchical relationship, and a near-synonym relationship; and a result data generation unit that generates result data in which the relationship between the document data connected by the generated semantic network is reflected in an image format, and displays the result data on the user terminal, wherein a user may visually check connection relationships according to the main concepts and keywords between the document data by means of the result data, and may explore document data related to the main concepts and keywords of the corresponding document data by selecting specific document data via the user terminal.

Description

Semantic network-based document relationship analysis device and method

[0001] The present invention relates to a semantic network-based document relationship analysis device and method, and more specifically, to a semantic network-based document relationship analysis device and method that utilizes artificial intelligence and natural language processing technology to analyze relationships between documents and thereby identify correlations between documents.

[0002] This invention was filed with support from the Gyeonggi Province and the Gyeonggi Province Economic and Science Promotion Agency's '2024 Global Startup Commercialization Support Project.'

[0003] In large-scale document databases, where countless documents are continuously created and updated, it is crucial to understand the relationships between these documents and explore semantically related documents. Existing document retrieval systems rely primarily on keyword-based retrieval techniques, often failing to capture the in-depth semantic relationships between documents. For example, while keyword retrieval is useful for finding documents containing specific words, it often fails to identify the topic or concept to which a document is related, or the semantic connections between two documents. Therefore, technologies are needed to extract key concepts and keywords within documents, build a semantic network, and then use this network to identify semantic relationships between documents. Such systems can contribute to the effective management and exploration of related documents by deeply analyzing the content of documents. For example, in research paper databases, semantic networks can be used to quickly search papers related to a specific topic, and in corporate document archives, they can facilitate the efficient management of work-related documents.

[0004] [Previous literature]

[0005] Registered patent 10-2215259

[0006] The present invention relates to a semantic network-based document relationship analysis device and method, and more specifically, to a semantic network-based document relationship analysis device and method that utilizes artificial intelligence and natural language processing technology to analyze relationships between documents and thereby identify correlations between documents.

[0007] A semantic network-based document relationship analysis device for analyzing relationships between documents based on a semantic network according to one embodiment of the present invention includes a document data receiving unit for receiving a plurality of document data from a user terminal, a document data analysis unit for analyzing the text content of each document data by utilizing a natural language processing technology and a text mining technique to extract main concepts and keywords of the document data, a semantic network generation unit for analyzing the main concepts and keywords extracted for each document data and generating a semantic network between the corresponding document data when it is determined that there is a synonymous relationship, a superior-subordinate relationship, or antonymous relationship between the main concepts and keywords, and a result data generation unit for generating result data in which the relationships between document data connected by the generated semantic network are reflected in the form of an image and displaying the result data on the user terminal, wherein the user can visually confirm the connection relationship according to the main concepts and keywords between the document data through the result data, and can select specific document data through the user terminal to search for document data related to the main concepts and keywords of the corresponding document data.

[0008] A group data generation unit that analyzes the main concepts and keywords of each document data, groups the document data in which the main concepts and keywords are semantically connected to each other, and generates group data, and a group data analysis unit that derives similarities and differences between the document data based on the main concepts and keywords of each document data within the generated group data, wherein the result data includes the contents derived by the group data analysis unit, and the user can visually confirm the contents through the result data.

[0009] A semantic network-based document relationship analysis device according to one embodiment of the present invention further includes a keyword receiving unit for receiving a keyword input by a user into a user terminal, and a first recommendation information transmitting unit for extracting document data for which a semantic network is generated with major concepts and keywords similar to the received keyword, and generating information about the extracted document data as first recommendation information and transmitting the information to the user terminal.

[0010] A semantic network-based document relationship analysis device according to one embodiment of the present invention further includes a selection information receiving unit that receives selection information including content about the selected specific document data from a user terminal when the user selects specific document data through a user terminal after the semantic network generating unit confirms result data, and a second recommendation information transmitting unit that extracts document data remaining in a semantic network including the document data included in the selection information, excluding the document data included in the selection information, and matches the extracted document data with key concepts and keywords extracted from the document data to generate second recommendation information and transmits the second recommendation information to the user terminal.

[0011] A method for analyzing a semantic network-based document relationship using a document relationship analysis device that analyzes relationships between documents based on a semantic network according to one embodiment of the present invention includes the steps of: a document data receiving unit receiving a plurality of document data from a user terminal; a document data analysis unit analyzing the text content of each document data using a natural language processing technology and a text mining technique to extract key concepts and keywords of the document data; a semantic network generation unit analyzing the key concepts and keywords extracted for each document data and generating a semantic network between the corresponding document data when it is determined that there is a synonymous relationship, a superior-subordinate relationship, or a synonymy relationship between the key concepts and keywords; and a result data generation unit generating result data in which the relationship between document data connected by the generated network is reflected in the form of an image and displaying the result data on the user terminal, wherein the user can visually confirm the connection relationship according to the key concepts and keywords between the document data through the result data, and can select specific document data through the user terminal to search for document data related to the key concepts and keywords of the corresponding document data.

[0012] The step of generating a semantic network includes a step in which the semantic network generation unit analyzes the main concepts and keywords of each document data, groups the document data in which the main concepts and keywords are semantically connected to each other to generate group data, and derives similarities and differences between the document data based on the main concepts and keywords of each document data within the generated group data, and the result data includes the similarities and differences between the document data derived in the step of generating the semantic network, and the user can visually confirm the corresponding contents through the result data.

[0013] The present invention utilizes natural language processing technology to analyze semantic relationships between document data to form a semantic network and provide detailed analysis information thereon to users, thereby enabling efficient exploration of large-scale document data and, furthermore, can be usefully used to efficiently manage large-scale document data.

[0014] FIG. 1 is a block diagram of a semantic network-based document relationship analysis system according to one embodiment of the present invention.

[0015] FIG. 2 is a block diagram of a document relationship analysis device according to one embodiment of the present invention.

[0016] Figure 3 is a flowchart of a document relationship analysis method according to one embodiment of the present invention.

[0017] A semantic network-based document relationship analysis device for analyzing relationships between documents based on a semantic network according to one embodiment of the present invention includes a document data receiving unit for receiving a plurality of document data from a user terminal, a document data analysis unit for analyzing the text content of each document data by utilizing a natural language processing technology and a text mining technique to extract main concepts and keywords of the document data, a semantic network generation unit for analyzing the main concepts and keywords extracted for each document data and generating a semantic network between the corresponding document data when it is determined that there is a synonymous relationship, a superior-subordinate relationship, or antonymous relationship between the main concepts and keywords, and a result data generation unit for generating result data in which the relationships between document data connected by the generated semantic network are reflected in the form of an image and displaying the result data on the user terminal, wherein the user can visually confirm the connection relationship according to the main concepts and keywords between the document data through the result data, and can select specific document data through the user terminal to search for document data related to the main concepts and keywords of the corresponding document data.

[0018] Below, with reference to the attached drawings, embodiments of the present invention are described in detail so that those skilled in the art can easily implement them. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In the drawings, irrelevant parts have been omitted for clarity of description, and similar reference numerals have been used throughout the specification to indicate similar elements.

[0019] Throughout the specification, when a part is said to be "connected" to another part, this includes not only cases where the parts are "directly connected," but also cases where the parts are "electrically connected" with other elements intervening. Furthermore, when a part is said to "include" a component, this does not exclude other components, but rather includes other components, unless otherwise specifically stated. The present invention will now be described in detail with reference to the accompanying drawings.

[0020] FIG. 1 is a block diagram of a semantic network-based document relationship analysis system (1000) according to one embodiment of the present invention.

[0021] Referring to FIG. 1, a semantic network-based document relationship analysis system (1000) may include a document relationship analysis device (200) connected to a user terminal (100) and a network (400).

[0022] The user terminal (100) may be a terminal used by a person who wants to quickly determine how documents are related through their content. For example, the user terminal (100) may be a terminal used by a person who analyzes and writes documents by referring to the relationships (associations) between documents.

[0023] The user terminal (100) may be a smartphone. However, the present invention is not limited thereto, and the user terminal (100) may include electronic devices such as general desktop computers, navigation systems, laptops, digital broadcasting terminals, PDAs (Personal Digital Assistants), PMPs (Portable Multimedia Players), tablet PCs, etc. The electronic device may have one or more general or special purpose processors, memory, storage, and / or networking components (wired or wireless).

[0024] The document relationship analysis device (200) receives multiple document data from a user terminal (100), analyzes the contents of the received document data, forms a semantic network of conceptually similar document data, and transmits the contents of the formed semantic network to the user terminal (100) so that the user can visually understand it. The document relationship analysis device (200) may be a server, and may be implemented in the form of an application within the user terminal (100). The document relationship analysis device (200) will be described in more detail with reference to FIGS. 2 and 3.

[0025] The communication method of the network (400) is not limited, and may include not only a communication method utilizing a communication network (e.g., a mobile communication network, a wired online network, a wireless online network, a broadcasting network) that the network (400) may include, but also short-range wireless communication between devices. For example, the network (400) may include one or more arbitrary networks (400) among networks (400) such as a personal area network (PAN), a local area network (LAN), a campus area network (CAN), a metropolitan area network (MAN), a wide area network (WAN), a broadband network (BBN), and an online network.

[0026] FIG. 2 is a block diagram of a document relationship analysis device (200) according to one embodiment of the present invention, and FIG. 3 is a flowchart of a document relationship analysis method according to one embodiment of the present invention.

[0027] Referring to FIGS. 2 and 3, the document relationship analysis device (200) may include a document data receiving unit (201), a document data analysis unit (202), a semantic network generating unit (203), a result data generating unit (204), a keyword receiving unit (205), a first recommendation information transmitting unit (206), a selection information receiving unit (207), and a second recommendation information transmitting unit (208).

[0028] The document data receiving unit (201) can receive multiple document data from the user terminal (100). (S11) The present invention can provide the user with the result data of analyzing the relationship (association) of the received multiple document data.

[0029] The document data analysis unit (202) can analyze the text content of each document data using natural language processing technology and text mining techniques to extract the main concepts and keywords of the document data. (S12) As an example of the present invention, the main concepts of the document data may be sentences expressing the subject of the document data, and the keywords may be key words related to the subject of the document data.

[0030] Natural language processing (NLP) technology, a branch of artificial intelligence, utilizes machine learning algorithms to process and interpret text within document data. For example, NLP is applied in email programs, which read and analyze the content of received messages to send appropriate reply messages. The document data analysis unit (202) of the present invention also utilizes NLP technology to analyze document data.

[0031] Text mining technology refers to an analysis and processing technology that extracts meaningful information for a specific purpose by using academic knowledge in linguistics, mathematics, statistics, and computer engineering.

[0032] The semantic network generation unit (203) analyzes the main concepts and keywords extracted for each document data, and if it is determined that there is a synonym relationship, a hierarchical relationship, or a synonym relationship between the main concepts and keywords, it can generate a semantic network between the corresponding document data (S13). A semantic network is a form of knowledge representation, and can be defined as a graph that represents nodes (verices) representing concepts and a semantic relationship between concepts, and the semantic network generation unit (203) of the present invention analyzes the main concepts and keywords between such document data, and if it is determined that there is a meaningful relationship (synonym, hierarchical relationship, or a synonym relationship) between the main concepts and keywords, it can form the corresponding document data into a semantic network.

[0033] The semantic network generation unit (203) may include a group data generation unit and a group data analysis unit.

[0034] The group data generation unit can analyze the main concepts and keywords of each document data, and group the document data in which the main concepts and keywords are semantically connected to each other to generate group data. As mentioned above, the group data generation unit can group the document data in which the main concepts and keywords are semantically connected to each other, such as synonym relationships, superior relationships, and antonym relationships, to generate group data.

[0035] The group data analysis unit can derive similarities and differences between document data within the generated group data based on the key concepts and keywords within each document data. For example, if the key concepts and keywords between document data are hierarchical, the group data analysis unit can derive similarities and differences based on that relationship. If the key concepts and keywords between document data are synonyms, the group data analysis unit can derive similarities and differences based on that relationship.

[0036] The result data generation unit (204) can generate result data in which the relationship between document data connected by the generated semantic network is reflected in image format and display the result data on the user terminal (100). (S14)

[0037] The result data includes the content derived by the group data analysis department, and users can visually confirm the content through the result data.

[0038] The keyword receiving unit (205) can receive a keyword entered by the user into the user terminal (100).

[0039] The first recommendation information transmission unit (206) can extract document data from which a semantic network is generated with key concepts and keywords similar to the received keywords, and generate information about the extracted document data as first recommendation information and transmit it to the user terminal (100).

[0040] A user can input a specific keyword to analyze document data in his / her user terminal (100), and the first recommendation information transmission unit (206) can provide, to the user terminal (100), document data in which a semantic network is formed with major concepts and keywords similar to the keywords received by the keyword reception unit (205), as first recommendation information. That is, a document data network having a relationship (association) with the keyword input by the user is extracted, and information on a plurality of document data linked to the extracted network can be generated as first recommendation information and provided to the user terminal (100).

[0041] The selection information receiving unit (207) can receive selection information including the content of the selected specific document data from the user terminal (100) when the user selects specific document data through the user terminal (100) after checking the result data.

[0042] The second recommendation information transmission unit (208) can extract document data other than the document data included in the selection information from a semantic network that includes the document data included in the selection information, match the extracted document data with the main concepts and keywords extracted from the document data, and generate second recommendation information and transmit it to the user terminal (100).

[0043] After the user checks the result data in which the network between document data is visually expressed, if the user selects the specific document data through the user terminal (100) to learn more about the specific document data, the second recommendation information transmission unit (208) extracts the semantic network including the selected specific document data, extracts the remaining document data excluding the specific document data from the extracted semantic network, and matches the extracted document data with the main concepts and keywords extracted by the document data analysis unit (202) to generate the second recommendation information and transmit it to the user terminal (100).

[0044] In this way, the present invention utilizes natural language processing technology to analyze semantic relationships between document data to form a semantic network and provide detailed analysis information thereon to users, thereby enabling efficient exploration of large-scale document data and, furthermore, can be usefully used to efficiently manage large-scale document data.

[0045] The embodiments described above are provided for illustrative purposes only, and those skilled in the art will readily appreciate that the embodiments described above can be readily modified into other specific forms without altering the technical concepts or essential characteristics of the embodiments described above. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, components described as being single may be implemented in a distributed manner, and similarly, components described as being distributed may be implemented in a combined manner.

[0046] The scope of protection sought through this specification is indicated by the claims described below rather than by the detailed description, and should be interpreted to include all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts.

Claims

1. In a semantic network-based document relationship analysis device that analyzes relationships between documents based on a semantic network, A document data receiving unit that receives multiple document data from a user terminal; A document data analysis unit that analyzes the text content of each document data using natural language processing technology and text mining techniques to extract key concepts and keywords from the document data; A semantic network generation unit that analyzes the extracted main concepts and keywords for each document data and generates a semantic network between the document data if it is determined that there is a synonymous relationship, a superior-subordinate relationship, or a synonym relationship between the main concepts and keywords; and Includes a result data generation unit that generates result data in which the relationship between document data connected by the generated semantic network is reflected in an image format and displays the result data on the user terminal. A semantic network-based document relationship analysis device characterized in that a user can initially confirm the relationship between document data according to main concepts and keywords through the result data, and can select specific document data through a user terminal to explore document data related to the main concepts and keywords of the document data.

2. In paragraph 1, The above semantic network generation unit includes a group data generation unit that analyzes the main concepts and keywords of each document data, groups document data in which the main concepts and keywords are semantically connected to each other, and generates group data, and a group data analysis unit that derives similarities and differences between document data based on the main concepts and keywords of each document data within the generated group data. A semantic network-based document relationship analysis device characterized in that the above result data includes content derived by the group data analysis unit, and a user can visually confirm the content through the result data.

3. In paragraph 2, A keyword receiving unit that receives a keyword entered by a user into a user terminal; and A semantic network-based document relationship analysis device further comprising a first recommendation information transmission unit that extracts document data generated by a semantic network with key concepts and keywords similar to received keywords, generates information about the extracted document data as first recommendation information, and transmits the information to the user terminal.

4. In paragraph 3, A selection information receiving unit that receives selection information including the contents of the selected specific document data from the user terminal after the user selects specific document data through the user terminal after checking the result data; and A semantic network-based document relationship analysis device further comprising a second recommendation information transmission unit that extracts document data excluding document data included in the selection information from a semantic network including document data included in the selection information, matches the extracted document data with key concepts and keywords extracted from the document data to generate second recommendation information, and transmits the second recommendation information to the user terminal.

5. A method for analyzing document relationships based on a semantic network using a document relationship analysis device that analyzes relationships between documents based on a semantic network. A step in which a document data receiving unit receives a plurality of document data from a user terminal; A step in which the document data analysis department analyzes the text content of each document data using natural language processing technology and text mining techniques to extract key concepts and keywords from the document data; A step of generating a semantic network between the document data when the semantic network generation unit analyzes the extracted main concepts and keywords for each document data and determines that there is a synonymous relationship, a superior-subordinate relationship, or a synonym relationship between the main concepts and keywords; and A step of generating result data in which the relationship between document data connected by the network generated by the result data generation unit is reflected in the form of an image and displaying the result data on the user terminal is included. A semantic network-based document relationship analysis method characterized in that a user can initially confirm the relationship between document data according to main concepts and keywords through the result data, and can select specific document data through a user terminal to explore document data related to the main concepts and keywords of the document data.

6. In paragraph 5, The step of generating a semantic network includes a step of the semantic network generation unit analyzing the main concepts and keywords of each document data, grouping the document data in which the main concepts and keywords are semantically connected to each other to generate group data, and a step of deriving similarities and differences between the document data based on the main concepts and keywords of each document data within the generated group data. A semantic network-based document relationship analysis method, characterized in that the result data includes similarities and differences between document data derived in the step of generating the semantic network, and a user can visually confirm the content through the result data.

Images