Lu xun theme intelligent interaction body system and interaction method based on large language model

Abstract

The application discloses a Lu Xun theme intelligent interaction body system and an interaction method based on a large language model, relies on a wisdom spectrum clear language large language model to construct an intelligent body with answering Lu Xun related problems as a core, realizes intelligent interaction with a user through exclusive configuration rules, a knowledge base calling mechanism, style and tone adaptation logic. The system presets exclusive response strategies for multiple types of problems, supports content output in Lu Xun's style, knowledge base information retrieval, learning encouragement and guidance, solves the problems that existing intelligent interaction bodies lack exclusive theme depth adaptation, style and tone expression is single, and response strategies are not targeted, and realizes precise, character characteristic fitting and multi-dimensional intelligent interaction effects under the Lu Xun theme. The interaction body system and method of the application only carry out interaction around the Lu Xun theme, the configuration rules and the response logic strictly match the Lu Xun related characteristics and requirements, there is no redundant function, the interaction experience is in line with the character setting and has the additional value of knowledge dissemination and guided learning.

Description

Technical Field

[0001] This invention relates to the field of intelligent interactive technology, specifically to an intelligent interactive system and interaction method based on a large language model for Lu Xun. Background Technology

[0002] With the development of large language model technology, various intelligent interactive entities are widely used in scenarios such as knowledge answering and intelligent dialogue. However, most existing intelligent interactive entities are general-purpose entities, lacking in-depth adaptation design for specific characters and themes, resulting in the following problems: First, the interaction scope of general-purpose intelligent interactive entities is not clearly defined, and the answers for specific characters lack professionalism and accuracy; second, the output content of existing interactive entities lacks exclusive style and tone adaptation, failing to match the expression characteristics of specific characters, and the interactive experience lacks a sense of immersion; third, the response strategies of existing interactive entities are relatively simple, lacking exclusive response rules for specific questions, and the tone of expression is fixed, unable to be flexibly adjusted according to the emotional attributes of the question; fourth, existing intelligent interactive entities only focus on question answering, lacking learning guidance functions for specific knowledge, and failing to realize the added value of knowledge dissemination.

[0003] In response to the demand for intelligent interaction on the theme of Lu Xun, there is currently no intelligent interactive body that is specifically built on a specific large language model, uses Lu Xun's tone and style as the core expressive features, and has a preset exclusive response strategy. General interactive bodies cannot meet users' needs for precise, character-appropriate, and multi-dimensional interaction on the theme of Lu Xun. Therefore, there is an urgent need for an intelligent interactive body system and interaction method that is specifically designed for the theme of Lu Xun. Summary of the Invention

[0004] The purpose of this invention is to provide a Lu Xun-themed intelligent interactive system and interaction method based on a large language model. This addresses the problems of existing general-purpose intelligent interactive systems that lack professionalism in answering Lu Xun-related questions, have monotonous stylistic and tonal expressions, lack targeted response strategies, and lack learning guidance functions. By relying on the Zhipu Qingyan large language model, this invention constructs an intelligent interactive system specifically for Lu Xun themes, achieving accurate answers to Lu Xun-related questions, tone and style expressions that match Lu Xun's characteristics, targeted response strategies and tone adjustments, and adding the added value of guiding users to learn about Lu Xun-related articles.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A Lu Xun-themed intelligent interactive system based on a large language model is proposed. Built upon the Zhipu Qingyan large language model, the system aims to answer various user questions about Lu Xun. The system includes a configuration module, a knowledge base module, a tone adaptation module, a voice adjustment module, and a response triggering module. The configuration module pre-stores Lu Xun-themed interactive configuration information, limiting the core interaction scope of the intelligent interactive system to all Lu Xun-related questions, and ensuring all interactive output is expressed in Lu Xun's voice. The knowledge base module stores comprehensive knowledge information related to Lu Xun, including his life experiences, works, famous quotes, social background, and ideological viewpoints, and supports real-time retrieval of information from the knowledge base for content output during interaction. The tone adaptation module incorporates Lu Xun's writings... The system employs a language model that extracts sentence structure, expression style, and lexical features from Lu Xun's writings and transforms them into language generation rules. This ensures that all responses from the intelligent interactive system conform to Lu Xun's stylistic characteristics. The tone adjustment module pre-sets tone expression logic for different emotional tendencies, matching the output tone according to the emotional attributes of the user's question. The response trigger module pre-stores specific response instructions for particular questions, automatically triggering the output of corresponding specific response content when a user raises a pre-set specific question. Furthermore, the system includes a learning guidance module that, during user interaction, outputs timely guidance content encouraging users to learn about Lu Xun's related articles. This learning guidance content is tailored to Lu Xun's tone and stylistic features, and remains within the interactive context of the Lu Xun-related questions.

[0006] Furthermore, the response triggering module pre-stores the first type of specific question as "Who are you?". When this question is detected, it automatically triggers a preset, fixed, and exclusive response without any additions, deletions, or modifications. The pre-stored second type of specific question is "What floor is your house on?". When this question and its synonyms are detected, it automatically triggers a preset, fixed, and exclusive response, matched with a lighthearted tone. The pre-stored third type of specific question is "Your famous quotes and sayings?". When this question and similar expressions are detected, it triggers a response logic that combines knowledge base excerpts with a humorous tone. The pre-stored fourth type of specific question is "Why did your family fall into decline?". When this question is detected, it triggers a response logic that combines knowledge base information retrieval with a sad tone.

[0007] Furthermore, the Lu Xun style language model of the tone adaptation module only extracts and trains features from Lu Xun's original texts such as classic works, manuscripts, and letters, without incorporating the stylistic features of other authors, thus ensuring the uniqueness and accuracy of Lu Xun's tone expression.

[0008] Furthermore, the tone adjustment module presets three core tones: a conventional answer tone, a humorous tone, and a sad tone. Each tone is customized in accordance with the stylistic features of Lu Xun's writings and automatically switches according to the emotional attributes of the user's question without human intervention.

[0009] Furthermore, the learning guidance module's learning encouragement content output follows the principle of scenario adaptation, only outputting in a timely manner after the core question has been answered, combined with the current interactive topic, without interfering with the answering of the core question, and the content conforms to Lu Xun's tone and stylistic characteristics.

[0010] Further, the process includes the following steps: Step S1, system initialization, loading preset rules and information for various modules; Step S2, receiving user questions and performing semantic recognition; Step S3, if the question is not related to Lu Xun, expressing in Lu Xun's tone that it cannot be answered and ending the interaction; Step S4, if the question is related to Lu Xun, determining whether it is a pre-stored specific question; Step S5, if it is a specific question, triggering the corresponding exclusive response logic to complete the content output; Step S6, if it is not a specific question, retrieving knowledge base information and combining Lu Xun's tone and corresponding intonation to complete the content output; Step S7, outputting learning encouragement content in a timely manner based on the interactive topic; Step S8, ending the current interaction and waiting for new question input or entering standby mode.

[0011] Furthermore, the judgment of a specific question in step S4 includes two methods: exact matching and synonym matching, to ensure accurate answers to the same question expressed by the user in different ways.

[0012] Furthermore, the learning guidance content output in step S7 is an optional step; if there is no suitable scenario, it will directly proceed to step S8.

[0013] This invention provides a Lu Xun-themed intelligent interactive system and interaction method based on a large language model, which has the following beneficial effects: Compared with the prior art, the present invention has the following advantages: 1. This invention relies on the Zhipu Qingyan big language model to construct an intelligent interactive body with a unique Lu Xun theme, strictly limiting the scope of interaction to Lu Xun-related questions, thus solving the problem of the lack of professionalism in the answers of existing general-purpose interactive bodies. By storing all relevant knowledge about Lu Xun through the knowledge base module, it achieves accurate answers to Lu Xun-related questions.

[0014] This invention sets up a tone adaptation module to construct a unique Lu Xun style language model, so that all output content of the intelligent interactive body conforms to Lu Xun's style and tone characteristics, greatly improving the sense of immersion in the interaction and solving the problem of monotonous style and tone expression in existing interactive bodies.

[0015] This invention includes a response trigger module that pre-stores exclusive response instructions for various specific questions and designs personalized response logic for different specific questions. It also includes a tone adjustment module that matches customized tones such as humor, sadness, and routine according to the emotional attributes of the question, thus achieving targeted response strategies and flexible tone expression.

[0016] This invention features a learning guidance module that outputs encouraging content in the style of Lu Xun in a timely manner after the user completes the question-and-answer session, thus adding value by guiding users to learn about Lu Xun's articles. This distinguishes it from existing intelligent interactive systems that only focus on answering questions.

[0017] The intelligent interactive system and method of this invention are designed with all configuration rules and response logic strictly around the theme of Lu Xun, with no redundant functions. The exclusive response content for all specific questions strictly follows the preset requirements without any additions, deletions or modifications, ensuring a high degree of fit between the interactive body and the characteristics of Lu Xun. Detailed Implementation

[0018] This embodiment provides a Lu Xun-themed intelligent interactive system based on a large language model. It is built on the Zhipu Qingyan large language model and its core purpose is to answer users' various questions about Lu Xun. The system hardware adopts a conventional intelligent interactive terminal, including a processor, a memory, and input / output devices. The memory stores the various module programs and data of the above system. The processor executes the programs to realize the various functions of the modules. The input / output devices realize the input of user questions and the output of intelligent interactive content.

[0019] The system includes a configuration module, a knowledge base module, a tone adaptation module, a tone adjustment module, a response triggering module, and a learning guidance module. The specific functions of each module are implemented as follows: 1. Configuration Module: The memory stores exclusive interactive configuration information for the Lu Xun theme. The configuration information clearly defines the core interaction scope of the intelligent interactive entity as only all questions related to Lu Xun, and all interactive output content must be expressed in Lu Xun's voice. This configuration information is the basic configuration of the system and is always effective after the system is initialized. It cannot be manually modified.

[0020] Knowledge Base Module: A database of knowledge about Lu Xun is established in the storage device. The stored information includes Lu Xun's life experience (birth year 1881, native place of Shaoxing, Zhejiang, his experience of abandoning medicine for literature, and his family's decline), classic works (such as "Diary of a Madman"), famous quotes, social background, ideological viewpoints, and the connotation of his articles, etc. All information in the database has been strictly verified to ensure authenticity and accuracy. The processor can retrieve information from the database in real time based on user questions during the interaction process and output the content.

[0021] The tone adaptation module embeds a language model featuring Lu Xun's stylistic characteristics into the processor. The model's construction process is as follows: First, it collects original text data from Lu Xun's classic works, manuscripts, and letters, such as "Diary of a Madman" and "The True Story of Ah Q." Then, it uses natural language processing technology to segment, parse, and extract the style of the original text data, extracting commonly used vocabulary types (such as words with era characteristics and critical words), sentence structures (such as the usage characteristics of rhetorical questions and declarative sentences), expression logic, and first-person perspective features in Lu Xun's style. The extracted features are then transformed into quantifiable language generation rules and embedded into the generation logic of the Zhipu Qingyan Big Language Model. This ensures that the intelligent interactive entity strictly follows these rules when generating any response content, guaranteeing that the style of the output content is highly consistent with Lu Xun's style. Furthermore, this model is only trained for Lu Xun's style and does not incorporate features from other authors.

[0022] The tone adjustment module: The processor pre-defines the expression logic of three core tones, each customized in accordance with the stylistic characteristics of Lu Xun's writings: a standard answer tone as the basic tone, suitable for routine questions, expressing objectivity and rigor; a humorous tone suitable for cheerful questions, achieving humor through lighthearted vocabulary and sentence structure adjustments in the Lu Xun style, without deviating from Lu Xun's aesthetic; and a sorrowful tone suitable for questions related to Lu Xun's tragic experiences, expressing sadness through emotional vocabulary and soothing sentence structures. The processor can determine the emotional attribute of the user's question through semantic recognition technology and automatically match the corresponding tone, without manual intervention.

[0023] Response triggering module: The memory pre-stores dedicated response instructions for four specific types of questions. The processor detects the user's question using semantic recognition technology and triggers the corresponding response. Once "Who are you?" is detected, preset fixed content is automatically output without any additions, deletions, or modifications. If the system detects "What floor is your house on?" and related expressions, it will automatically output preset fixed content and match it with a relaxed tone. Once "your famous quotes" and similar expressions are detected, we first accurately extract famous quotes from Lu Xun from the knowledge base, and then output them with a humorous tone. The system detects the question "Why did your family fall into poverty?" and first retrieves relevant biographical information from the knowledge base. Then, it matches the question with a sad tone and outputs the answer in the first person, in a style consistent with Lu Xun's writing style.

[0024] Learning guidance module: The processor stores learning encouragement output rules. After answering the user's core questions, it outputs encouraging content in a timely manner based on the current interactive topic. For example, if a user asks a question related to "Diary of a Madman", after answering, it outputs content in the style of Lu Xun such as "This short article is simple, but it contains reflections on the world. You can read it carefully to understand its profound meaning." If there is no suitable scenario for the current topic, no content will be output.

[0025] This embodiment also provides an interaction method based on the above system, the specific steps of which are as follows: Step S1: System initialization. The user starts the intelligent interactive terminal and launches the Lu Xun-themed intelligent interactive body based on the Zhipu Qingyan big language model. The processor loads all preset rules, data and instructions from the configuration module, knowledge base module, tone adaptation module, tone adjustment module, response trigger module and learning guidance module.

[0026] In step S2, the user inputs a question through an input device (keyboard, voice input, etc.). After the intelligent interactive body receives the question, the processor performs semantic recognition on the question.

[0027] Step S3: If the semantic recognition result is a question not related to Lu Xun (such as asking about other authors, irrelevant life questions, etc.), the processor generates unanswerable content that matches Lu Xun's tone through the tone adaptation module, outputs it through the output device, and ends the interaction.

[0028] Step S4: If the semantic recognition result is a question related to Lu Xun, the processor further determines whether the question is a specific question pre-stored in the response trigger module. The determination methods include exact matching and synonym matching.

[0029] In step S5, if a specific problem is identified, the processor automatically triggers the corresponding dedicated response logic according to the preset instructions of the response trigger module to complete the content output. For example, if "Who are you?" is detected, preset fixed content is directly output.

[0030] In step S6, if the question is determined to be non-specific (such as asking about the content of a certain article by Lu Xun, the details of a certain experience, etc.), the processor first retrieves relevant knowledge information from the knowledge base module, then converts the information into content that conforms to Lu Xun's writing style through the tone adaptation module, and at the same time matches the corresponding tone according to the emotional attributes of the question through the tone adjustment module, and completes the content output through the output device.

[0031] Step S7: The processor determines whether there is a suitable learning guidance scenario for the current interactive topic. If there is, the learning guidance module outputs learning encouragement content that matches Lu Xun's style; if not, proceed directly to step S8.

[0032] Step S8: This interaction ends. The intelligent interactive object waits for the user to input the next question. If no new question is input within 5 minutes, the system enters standby mode to save terminal resources.

[0033] In this embodiment, all output content of the intelligent interactive body strictly follows Lu Xun's tone and stylistic features. The exclusive response content for all specific questions is completely consistent with the preset content, without any additions, deletions or modifications. The information retrieval from the knowledge base is guaranteed to be authentic and accurate. The tone adjustment and learning guidance are all in line with Lu Xun's theme and interactive context, realizing a precise, character-specific, and multi-dimensional intelligent interactive effect under the theme of Lu Xun.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A Lu Xun-themed intelligent interactive system based on a large language model, characterized in that: Based on the Zhipu Qingyan big language model, an intelligent interactive system is built with the core purpose of answering users' various questions about Lu Xun. The system includes a configuration module, a knowledge base module, a tone adaptation module, a tone adjustment module, and a response triggering module. The configuration module pre-stores Lu Xun-themed exclusive interactive configuration information. The configuration information limits the core interaction scope of the intelligent interactive body to all questions related to Lu Xun, and all interactive output content is expressed in Lu Xun's voice. The knowledge base module stores all knowledge information related to Lu Xun, including Lu Xun's life experiences, works, famous quotes, social background, and ideological viewpoints. It also supports intelligent interactive objects to retrieve information from the knowledge base in real time during the interaction process for content output. The tone adaptation module has a built-in language model of Lu Xun's stylistic features. This model extracts the sentence structure, expression style, and vocabulary features of Lu Xun's writing style and transforms them into language generation rules, so that all the responses of the intelligent interactive body conform to the stylistic expression characteristics of Lu Xun. The tone adjustment module is preset with tone expression logic for different emotional tendencies, and can match the corresponding output tone according to the emotional attributes of the user's question; The response triggering module pre-stores exclusive response instructions for specific questions. When it detects that a user has asked a pre-set specific question, it automatically triggers the output of the corresponding exclusive response content. The system also includes a learning guidance module, which can output guidance content encouraging users to learn about Lu Xun's articles during the interaction with the user. The output of the learning guidance content is combined with Lu Xun's tone and stylistic features, and does not deviate from the interactive context of Lu Xun-related questions.

2. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The first type of specific question pre-stored in the response triggering module is "Who are you?" When the response triggering module detects that the user inputs this question, it will automatically trigger the preset exclusive response content, which is "I am Zhou Shuren, courtesy name Yu Cai, born in Shaoxing, Zhejiang in 1881." The world is more familiar with my pen name 'Lu Xun', which I started using when I published 'Diary of a Madman' in 1918; Despite the chaos of the times, he still cherished a pure heart, wishing to illuminate the darkness with his words, guide the youth forward, and jointly build the future of the nation. When the response trigger module is activated, it will simultaneously use a tone adaptation module to ensure that the tone and rhythm of the output content match Lu Xun's expressive characteristics, without any additions, deletions, or modifications to the content.

3. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The second type of specific question pre-stored in the response triggering module is "What floor is your house on?" When the response triggering module detects that the user inputs this question, including synonyms of the question, it will automatically trigger the preset exclusive response content, which is "I have moved many times". When the response triggering module triggers this exclusive response content, it simultaneously matches a relaxed tone through the tone adjustment module, and the content output format maintains the original preset expression, including the complete presentation of punctuation marks and modal particles.

4. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The third type of specific question pre-stored in the response triggering module is "your famous quotes". When the response triggering module detects that the user has entered this question, including similar expressions, it will trigger the response logic that combines knowledge base retrieval with a humorous tone. First, the knowledge base module accurately extracts famous quotes from Lu Xun's stored knowledge information according to preset extraction rules. The extracted content ensures the accuracy and completeness of the original sentence, without tampering, deletion or rewriting. Then, the tone adjustment module matches the extracted famous quotes with a humorous tone. At the same time, it combines the Lu Xun style rules of the tone adaptation module to form a response that combines knowledge accuracy, humor, and Lu Xun's style.

5. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The fourth type of specific question pre-stored in the response triggering module is "Why did your family fall into decline?" When the response triggering module detects that the user inputs this question, it will trigger a special response logic with a sad tone. First, the knowledge base module retrieves biographical information related to Lu Xun's family's decline. The retrieved information is guaranteed to be consistent with Lu Xun's real experiences. Then, the tone adjustment module matches the sad expression tone. At the same time, the tone adaptation module transforms the retrieved information into expression content that conforms to Lu Xun's writing style. The response content, while ensuring the authenticity of the information, adjusts the vocabulary and sentence structure to fit the sad emotional tendency, and is expressed from Lu Xun's first-person perspective throughout.

6. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The process of constructing the Lu Xun style language model for the tone adaptation module is as follows: First, original text data such as Lu Xun's classic works, manuscripts, and letters are collected. The original text data is segmented, parsed, and style extracted to extract commonly used vocabulary types, sentence structures, expression logic, and perspective features in Lu Xun's style. Then, the extracted features are transformed into quantifiable language generation rules and embedded into the generation logic of the Zhipu Qingyan Big Language Model. This ensures that the intelligent interactive entity follows these language generation rules when generating any response content, guaranteeing that the style of the output content is highly consistent with Lu Xun's. Furthermore, this language model is only trained and adapted for Lu Xun's style and does not integrate the style features of other authors, ensuring the uniqueness and accuracy of Lu Xun's tone expression.

7. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The tone adjustment module has three core tone expression logics: humorous tone, sad tone, and conventional answering tone. Each tone is customized in combination with Lu Xun's stylistic features. The standard answer tone is the basic tone of the intelligent interactive body, which is suitable for users to ask routine questions related to Lu Xun, ensuring that the expression is objective, rigorous and in line with Lu Xun's tone; the humorous tone is suitable for users to ask questions about happiness, and achieves humorous expression by adjusting the vocabulary and sentence structure under the characteristics of Lu Xun's writing style, without deviating from Lu Xun's expression style. The sad tone is suitable for questions related to Lu Xun's sad experiences. It achieves the expression of sadness through the emotional adjustment of vocabulary and the design of soothing sentence structure. Moreover, the switching of various tones is automatically identified and matched by the tone adjustment module according to the emotional attributes of the user's question, without human intervention.

8. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 1, characterized in that: The learning guidance module's output of learning encouragement content follows the principle of scenario adaptation. It only outputs encouragement content in a timely manner, combined with the current interactive topic, after the intelligent interactive entity has answered the user's questions related to Lu Xun. The encouragement content is expressed in Lu Xun's tone and style. For example, in combination with the user's question about a certain article by Lu Xun, the output encourages the user to read the article in depth and learn the ideas behind it. In combination with the user's question about a famous quote by Lu Xun, the output encourages the user to understand the meaning of the quote and learn Lu Xun's thoughts. Moreover, the learning guidance content is not lengthy or out of place in the current interactive context. It only serves to guide learning and does not interfere with the answering of the core questions.

9. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in any one of claims 1-8, characterized in that: Includes the following steps: Step S1: System initialization. Based on the Zhipu Qingyan big language model, the Lu Xun-themed intelligent interactive body is launched, loading the exclusive interactive configuration information in the configuration module, the Lu Xun-related knowledge information in the knowledge base module, the Lu Xun style language generation rules in the tone adaptation module, the various tone expression logics in the tone adjustment module, the exclusive response instructions for specific questions in the response trigger module, and the learning encouragement output rules in the learning guidance module. Step S2: User inputs question. The intelligent interactive body receives the question input by the user, performs semantic recognition on the question, and determines whether the question is related to Lu Xun. Step S3: Determine if the semantic recognition result is not related to Lu Xun. If the semantic recognition result is not related to Lu Xun, the intelligent interactive entity will express the content that cannot answer the question in Lu Xun's tone and style, and end the interaction. Step S4: Determine if the semantic recognition result is a question related to Lu Xun. If the semantic recognition result is a question related to Lu Xun, the intelligent interactive body will further determine whether the question is a specific question pre-stored in the response trigger module. Step S5, Specific Question Response: If it is determined to be a specific question, the intelligent interactive body will automatically trigger the corresponding exclusive response logic according to the preset instructions of the response trigger module, and complete the exclusive content output. During the output process, the corresponding tone and Lu Xun's tone will be matched synchronously. Step S6, non-specific question response: If it is determined to be a non-specific question, the intelligent interactive body first retrieves Lu Xun knowledge information related to the question from the knowledge base module, and then the tone adaptation module converts the retrieved information into content that conforms to the characteristics of Lu Xun's writing style. At the same time, the tone adjustment module matches the corresponding tone according to the emotional attributes of the question to complete the content output. Step S7, Learning Guidance Output: After completing the core question answers in Step S5 or Step S6, the learning guidance module outputs learning encouragement content that conforms to Lu Xun's tone and style, based on the current interactive topic, to guide users to learn about Lu Xun's related articles. Step S8: This interaction ends. The intelligent interactive object waits for the user's next question input. If no new question is input within a preset time, the system enters standby mode.

10. The intelligent interactive system and interaction method based on a large language model for Lu Xun as described in claim 9, characterized in that: In step S4, the judgment of a specific question includes two methods: exact matching and synonym matching. Exact matching means that the question entered by the user is completely consistent with the specific question pre-stored in the response trigger module. Synonymous matching means that the question entered by the user is semantically identical to the pre-stored specific question, only the expression is different. In step S5, the exclusive response logic for specific questions strictly follows the preset rules. For example, the response logic for the question "Your famous quotes" requires accurate extraction from the knowledge base first, followed by adaptation to a humorous tone. The response logic for the question "Why did your family fall into decline?" requires information retrieval from the knowledge base first, followed by adaptation to a sad tone. Furthermore, the exclusive response content for all specific questions is guaranteed to be consistent with the preset content, without any additions, deletions, or modifications. The learning guidance content output in step S7 is an optional step. If there is no suitable learning guidance scenario for the current interactive topic, the learning guidance module will not output any content and will directly proceed to step S8. Throughout the entire interaction process, all output content of the intelligent interactive body strictly follows the Lu Xun style rules of the tone adaptation module to ensure that the entire process is expressed in Lu Xun's tone and there is no other style of content output.