Network novel auxiliary reading system based on artificial intelligence

Abstract

The invention relates to the technical field of auxiliary reading, and particularly discloses a network novel auxiliary reading system based on artificial intelligence, which comprises a file import module, an abstract generation module, a character relationship prediction module, an event generation module and a chapter reading guide module, the file import module processes the imported network novel file; the abstract generation module is used for segmenting the processed network novel file into different chapters and generating abstract content corresponding to each chapter; the character relationship prediction module predicts the character relationship of each chapter according to the abstract content of each chapter; the event generation module generates event description information of each chapter according to the abstract content of each chapter; and the chapter reading guide module constructs a character relation graph of each chapter according to the character relation of each chapter, and displays the abstract content, the character relation graph and the event description information of each chapter. The novel content can be automatically analyzed and presented to a reader in a visual mode, the reader is helped to remember the character relation and events of the novel, and the reading efficiency and the understanding depth are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of auxiliary reading, and in particular to an auxiliary reading system for online novels based on artificial intelligence. Background Art

[0002] With the development of the Internet and the popularity of digital reading, online novels have become a popular form of literature. Especially in China, online novels have attracted a large number of readers with their rich content, unique narrative style and wide range of themes. However, online novels tend to have long lengths and complex plots, including a large number of characters and events, which may make it difficult for readers to follow the characters' relationships and development.

[0003] At present, most online reading platforms provide relatively simple functions, mainly focusing on basic text display, chapter navigation, etc., but lack the functions of in-depth analysis of novel content and auxiliary understanding. For example, when readers encounter unfamiliar characters or events, they may need to find relevant information by themselves, which not only interrupts the continuity of reading, but also increases the difficulty of reading.

[0004] In addition, online novels usually have the following characteristics: Large number of characters: A novel may contain dozens or even hundreds of characters, and the relationships between these characters are complex.

[0005] Rich Plot: The storyline spans multiple time periods and involves a variety of events and turning points.

[0006] Complex cultural background: Many novels incorporate rich historical and cultural elements, which may be difficult for non-native readers to understand.

[0007] Therefore, existing reading tools have certain limitations in helping readers understand the content of novels. Although there are some tentative solutions on the market, such as providing auxiliary information through simple keyword searches or chapter-based indexes, these methods are often not smart enough and cannot provide a comprehensive view of character relationships and event developments. Summary of the invention

[0008] In view of the defects and shortcomings in the prior art, the present invention provides an artificial intelligence-based online novel auxiliary reading system that can automatically analyze the content of the novel, extract key information, such as character relationships, important events, etc., and present it to readers in an intuitive manner to improve reading efficiency and depth of understanding.

[0009] As a first aspect of the present invention, there is provided an artificial intelligence-based online novel auxiliary reading system, the artificial intelligence-based online novel auxiliary reading system comprising a file import module, a summary generation module, a character relationship prediction module, an event generation module and a chapter guide module; wherein: The file import module is used to import an online novel file and process the online novel file to obtain a processed online novel file; The summary generation module is used to divide the processed online novel file into different chapters and generate summary content corresponding to each chapter; The character relationship prediction module is used to predict the character relationship of each chapter based on the summary content of each chapter; The event generation module is used to generate event description information for each chapter according to the summary content of each chapter; The chapter introduction module is used to construct a character relationship diagram for each chapter based on the character relationship in each chapter, and to display the summary content, character relationship diagram and event description information of each chapter.

[0010] Furthermore, the file import module is specifically used to convert the character encoding of the online novel file into UTF-8 encoding to obtain a converted online novel file; then clear the meaningless characters in the converted online novel file to obtain a cleared online novel file; finally, unify the format of the cleared online novel file to obtain an online novel file with a unified format.

[0011] Furthermore, the file import module uses the chardet tool of Python to automatically detect and convert the character encoding of the online novel file to unify the character encoding of the online novel file into UTF-8 encoding.

[0012] Furthermore, the file import module uses regular expressions to remove meaningless characters in the converted online novel file to obtain a cleaned online novel file; wherein the meaningless characters include any one or more of noise characters, punctuation marks and HTML tags.

[0013] Furthermore, the file import module uses a format standardization tool to unify the line spacing and paragraph format of the cleared online novel file to obtain an online novel file with a unified format.

[0014] Furthermore, the summary generation module is specifically used to fine-tune the large language model of the Transformer architecture on the historical data set of online novel files to obtain a fine-tuned large language model; and then use the fine-tuned large language model to generate corresponding summary content for each chapter.

[0015] Furthermore, the character relationship prediction module is specifically used to, first, use a bidirectional LSTM model to encode the summary content of each chapter; secondly, use a CRF model to determine and extract the character names in the summary content of each chapter; then input the summary content of each chapter and any two character names extracted therefrom into a trained Bert model to predict the character relationship between any two character names; finally, convert any two character names in each chapter and the character relationship between them into triple information and store it in a database; wherein the triple information is in the form of: {character name 1, character relationship, character name 2}.

[0016] Furthermore, the chapter introduction module is specifically used to construct a character relationship diagram for each chapter using the Neo4j framework of the knowledge graph using the triple information in the database.

[0017] Furthermore, the event generation module is specifically used to first use a convolutional layer or a self-attention mechanism to capture the contextual information in the summary content of each chapter to extract all event trigger words in each chapter; then classify all events in each chapter; then identify the entity of each event in each chapter; and finally combine the type and entity of each event in each chapter into an event description information.

[0018] Furthermore, the chapter guide module is specifically used to display a series of event description information of each chapter in a timeline according to the extraction order of event trigger words in each chapter, wherein the series of event description information is in the form of: {event description 1->event description 2->event description 3->…}.

[0019] The artificial intelligence-based online novel auxiliary reading system provided by the present invention has the following advantages: it can automatically analyze the content of the novel, extract key information, such as character relationships, important events, etc., and present it to readers in an intuitive manner, helping readers remember the character relationships and events in the novel, avoiding the situation of wasting time reading previous chapters due to forgetting, so as to improve reading efficiency and depth of understanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present invention, but do not constitute a limitation of the present invention.

[0021] Figure 1 This is a structural block diagram of the artificial intelligence-based online novel auxiliary reading system provided by the present invention. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment 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 creative work should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] In this embodiment, a network novel auxiliary reading system based on artificial intelligence is provided. Figure 1 This is a structural block diagram of the network novel auxiliary reading system based on artificial intelligence provided by the present invention. Figure 1 As shown, the artificial intelligence-based online novel auxiliary reading system includes a file import module, a summary generation module, a character relationship prediction module, an event generation module and a chapter guide module; wherein, The file import module is used to import an online novel file and process the online novel file to obtain a processed online novel file; It should be noted that the file import module allows users to import online novel files of various formats into the online novel auxiliary reading system, including but not limited to files in txt and pdf formats. The implementation of this function ensures that users can easily integrate online novel files from different sources into the system.

[0026] The summary generation module is used to divide the processed online novel file into different chapters and generate summary content corresponding to each chapter; The character relationship prediction module is used to predict the character relationship of each chapter based on the summary content of each chapter; The event generation module is used to generate event description information for each chapter according to the summary content of each chapter; The chapter introduction module is used to construct a character relationship diagram for each chapter based on the character relationship in each chapter, and to display the summary content, character relationship diagram and event description information of each chapter.

[0027] Preferably, the file import module is specifically used to convert the character encoding of the online novel file into UTF-8 encoding to obtain a converted online novel file; then clear the meaningless characters in the converted online novel file to obtain a cleared online novel file; finally, unify the format of the cleared online novel file to obtain an online novel file with a unified format.

[0028] Preferably, the file import module uses Python's chardet tool to automatically detect and convert the character encoding of the online novel file to unify the character encoding of the online novel file into UTF-8 encoding, thereby avoiding parsing errors.

[0029] Preferably, the file import module uses regular expressions to remove meaningless characters in the converted online novel file to obtain a cleaned online novel file; wherein the meaningless characters include any one or more of noise characters, punctuation marks and HTML tags.

[0030] Preferably, the file import module uses a format standardization tool to unify the line spacing and paragraph format of the cleared online novel file to obtain an online novel file with a unified format, thereby providing a consistent reading experience for users.

[0031] Preferably, the summary generation module is specifically used for, first, dividing the processed online novel file into different chapters according to structural features such as chapter titles and paragraph switches; second, fine-tuning a large language model (such as Lamma3) of the Transformer architecture open sourced on Hugging Face on a historical dataset of online novel files to obtain a fine-tuned large language model; and then using the fine-tuned large language model to generate corresponding summary content for each chapter.

[0032] Preferably, the character relationship prediction module is specifically used to, first, use a bidirectional LSTM model to encode the summary content of each chapter; secondly, use a CRF (conditional random field) model to determine and extract the character names in the summary content of each chapter; then input the summary content of each chapter and any two character names extracted therefrom into a trained Bert model to predict the character relationship between any two character names; finally, convert any two character names in each chapter and the character relationship between them into triple information and store it in a database; wherein the triple information is in the form of: {character name 1, character relationship, character name 2}.

[0033] It should be noted that the system will define the relationship between two characters, such as master and apprentice, lovers, mother and child, etc.

[0034] Preferably, the chapter guide module is specifically used to construct a character relationship diagram for each chapter using the triple information in the database using the Neo4j framework of the knowledge graph.

[0035] Preferably, the event generation module is specifically used to automatically identify and extract event information with practical significance from the summary content of each chapter, and the process is implemented using the DMCNN algorithm, as follows: First, DMCNN uses convolutional layers or self-attention mechanisms to capture the contextual information in the summary content of each chapter to extract all event trigger words in each chapter; then all events in each chapter are classified; then the entities or other words of each event in each chapter are identified; finally, the type, entity, other words and other information of each event in each chapter are combined into an event description information.

[0036] Preferably, the chapter guide module is specifically used to display a series of event description information of each chapter in a timeline according to the extraction order of event trigger words in each chapter, wherein the series of event description information is in the form of: {event description 1->event description 2->event description 3->…}.

[0037] An artificial intelligence-based online novel auxiliary reading system provided by an embodiment of the present invention can automatically analyze the content of the novel, extract key information, such as character relationships, important events, etc., and present it to readers in an intuitive manner, helping readers to remember the character relationships and events in the novel, avoiding the situation of wasting time reading previous chapters due to forgetting, thereby improving reading efficiency and depth of understanding.

[0038] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present invention, but the present invention is not limited thereto. For those of ordinary skill in the art, various modifications and improvements can be made without departing from the spirit and essence of the present invention, and these modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. An artificial intelligence-based online novel auxiliary reading system, characterized in that: The artificial intelligence-based online novel auxiliary reading system includes a file import module, a summary generation module, a character relationship prediction module, an event generation module and a chapter guide module; wherein, The file import module is used to import an online novel file and process the online novel file to obtain a processed online novel file; The summary generation module is used to divide the processed online novel file into different chapters and generate summary content corresponding to each chapter; The character relationship prediction module is used to predict the character relationship of each chapter based on the summary content of each chapter; The event generation module is used to generate event description information for each chapter according to the summary content of each chapter; The chapter introduction module is used to construct a character relationship diagram for each chapter based on the character relationship in each chapter, and to display the summary content, character relationship diagram and event description information of each chapter.

2. The artificial intelligence-based online novel auxiliary reading system according to claim 1 is characterized in that: The file import module is specifically used to convert the character encoding of the online novel file into UTF-8 encoding to obtain a converted online novel file; then clear the meaningless characters in the converted online novel file to obtain a cleared online novel file; finally, unify the format of the cleared online novel file to obtain an online novel file with a unified format.

3. The artificial intelligence-based online novel auxiliary reading system according to claim 2 is characterized in that: The file import module uses the chardet tool of Python to automatically detect and convert the character encoding of the online novel file to unify the character encoding of the online novel file into UTF-8 encoding.

4. The artificial intelligence-based online novel auxiliary reading system according to claim 2 is characterized in that: The file import module uses regular expressions to remove meaningless characters in the converted online novel file to obtain a cleaned online novel file; wherein the meaningless characters include any one or more of noise characters, punctuation marks and HTML tags.

5. The artificial intelligence-based online novel auxiliary reading system according to claim 2 is characterized in that: The file import module uses a format standardization tool to unify the line spacing and paragraph format of the cleaned online novel file to obtain an online novel file with a unified format.

6. The artificial intelligence-based online novel auxiliary reading system according to claim 1 is characterized in that: The summary generation module is specifically used to fine-tune the large language model of the Transformer architecture on the historical data set of online novel files to obtain a fine-tuned large language model; and then use the fine-tuned large language model to generate corresponding summary content for each chapter.

7. The artificial intelligence-based online novel auxiliary reading system according to claim 1 is characterized in that: The character relationship prediction module is specifically used to first use a bidirectional LSTM model to encode the summary content of each chapter; secondly, use a CRF model to determine and extract the character names in the summary content of each chapter; then input the summary content of each chapter and any two character names extracted therefrom into a trained Bert model to predict the character relationship between any two character names; finally, convert any two character names in each chapter and the character relationship between them into triple information and store it in a database; wherein the triple information is in the form of: {character name 1, character relationship, character name 2}.

8. The artificial intelligence-based online novel auxiliary reading system according to claim 7 is characterized in that: The chapter introduction module is specifically used to construct a character relationship diagram for each chapter using the Neo4j framework of the knowledge graph using the triple information in the database.

9. The artificial intelligence-based online novel auxiliary reading system according to claim 1 is characterized in that: The event generation module is specifically used to first use a convolutional layer or a self-attention mechanism to capture the contextual information in the summary content of each chapter to extract all event trigger words in each chapter; then classify all events in each chapter; then identify the entity of each event in each chapter; and finally combine the type and entity of each event in each chapter into an event description information.

10. The artificial intelligence-based online novel auxiliary reading system according to claim 9 is characterized in that: The chapter guide module is specifically used to display a series of event description information of each chapter in a timeline according to the extraction order of event trigger words in each chapter, wherein the series of event description information is in the form of: {event description 1->event description 2->event description 3->…}.

Images