Intelligent interaction system and method based on legal consultation

Abstract

The invention relates to the technical field of artificial intelligence, and discloses an intelligent interaction system and method based on legal consultation, and the system comprises a front-end user interface which supports text input, voice input and image input and has a multi-language switching function; the middle layer processing module comprises an emotion analysis module, an information integration module, a context management module and a response generation module and can dynamically adjust a processing strategy; and the rear-end legal knowledge base is used for storing laws and regulations, case analysis and common legal question answers and supporting real-time updating, version management and efficient retrieval. Through combination of a deep learning model and fine tuning of a special data set in the legal field, the system can accurately identify emotion features in user input, the accuracy not only improves the ability to understand legal terms, but also enhances the ability to process complex emotion expression, so that legal consultation is more humanized, and the efficiency is improved. The emotion change of the user can be responded in time, so that the satisfaction degree and the trust feeling of the user are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent interactive system and method based on legal consultation. Background Art

[0002] With the development of information technology, the legal consulting industry has gradually introduced emerging technologies such as artificial intelligence and machine learning to improve service efficiency and accessibility. Despite this, the existing legal consulting system still has many shortcomings, which restricts its effectiveness in practical applications. At present, many legal consulting systems rely on simple sentiment analysis tools and often only use sentiment dictionaries for sentiment judgment. This method cannot deeply understand the complex emotions in user input, especially in the legal field. Users' emotional expressions are often accompanied by specific legal backgrounds and contexts. Therefore, the system lacks accuracy and timeliness in identifying user emotions, resulting in the response to legal advice not fully reflecting the user's real needs, affecting the consulting effect.

[0003] Existing legal consulting platforms usually adopt a standardized response model and lack attention to individual differences of users. This lack of personalized service cannot meet the specific needs of users, especially when dealing with complex legal issues. Users want to obtain specific and in-depth legal guidance, while existing systems often provide unchanging advice, resulting in poor user experience and even reducing users' trust in legal services.

[0004] In the traditional legal consultation process, users need to input information through multiple channels, and the system usually needs to go through tedious data integration and analysis steps. This inefficient processing flow not only prolongs the user's waiting time, but also increases the risk of errors in information processing.

[0005] In view of the deficiencies of the prior art, the present invention provides an intelligent interactive system and method based on legal consultation. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides an intelligent interactive system and method based on legal consultation, which solves the problem that many legal consultation systems rely on simple sentiment analysis tools and often only use sentiment dictionaries for sentiment judgment, and this method cannot deeply understand the complex emotions in user input.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent interactive system based on legal consultation, comprising:

[0008] The front-end user interface supports text input, voice input, and image input, and has multi-language switching function;

[0009] The middle layer processing modules, including sentiment analysis module, information integration module, context management module and response generation module, can dynamically adjust the processing strategy;

[0010] The back-end legal knowledge base stores laws and regulations, case analysis and answers to common legal questions, and supports real-time updates, version management and efficient retrieval.

[0011] Preferably, the sentiment analysis module includes:

[0012] A deep learning model for receiving user input and performing emotion recognition. The model is self-learning and can be continuously optimized based on user feedback;

[0013] Sentiment dictionary, which annotates sentiment words related to the legal field, their sentiment categories and intensities, and supports the expansion of user-defined vocabulary.

[0014] Preferably, the sentiment recognition model is trained based on the BERT model and uses a special data set in the legal field to fine-tune the model, optimize the recognition ability of legal terms, and integrate multiple sentiment analysis algorithms to enhance the comprehensive judgment ability of the model.

[0015] Preferably, the multimodal user interface has a responsive design that can provide a consistent user interaction experience on different devices, and has adaptive layout and theme switching functions to meet the preferences of different users. Different devices include PCs, smartphones and tablets.

[0016] Preferably, the speech recognition module is combined with a legal terminology optimization algorithm to use an efficient speech recognition engine, and optimizes legal terminology for specific scenarios to improve recognition accuracy, and is capable of processing speech inputs with different accents and speaking speeds.

[0017] Preferably, the image processing module includes an optical character recognition (OCR) function that can extract text information from legal documents uploaded by users, supports processing of files in different formats such as PDF, JPEG, and PNG, and combines NLP technology to perform legal relevance analysis on the extracted content to improve information processing efficiency.

[0018] Preferably, the response generation module generates personalized legal advice based on user sentiment and input context, and adjusts the response strategy in real time in combination with historical consultation data and user feedback to ensure that the generated advice is highly relevant and accurate.

[0019] An intelligent interactive method based on legal consultation, comprising:

[0020] Collect user text, voice and image input, and use multimodal fusion technology to uniformly process user data in different input forms;

[0021] Preprocess the collected input, including cleaning and noise removal of text data, and transcribing and standardizing speech data to ensure data quality;

[0022] Use sentiment analysis models to identify user emotions, generate sentiment reports, and combine them with user input to provide real-time feedback on emotional status;

[0023] Integrate user input and sentiment reports to generate response suggestions, and use natural language generation (NLG) technology to output professional and easy-to-understand legal consulting content.

[0024] Preferably, the emotion recognition step includes:

[0025] Use sentiment dictionaries to extract sentiment features from user input and combine contextual information to identify changes in user emotional states;

[0026] A variety of sentiment analysis algorithms including sentiment polarity analysis and sentiment intensity assessment are applied to conduct a comprehensive analysis of user input to improve the accuracy of sentiment recognition.

[0027] Preferably, the step of generating a personalized response includes:

[0028] Generate legal advice content that meets user needs based on user emotions and input context, and optimize response strategies based on user feedback to ensure that the legal advice provided meets user expectations in terms of professionalism and personalization. At the same time, record users' historical consultation data to provide more accurate background information for subsequent interactions.

[0029] The present invention provides an intelligent interactive system and method based on legal consultation, which has the following beneficial effects:

[0030] 1. By combining the deep learning model with fine-tuning of a dedicated legal data set, the system can accurately identify the emotional features in user input. This accuracy not only improves the ability to understand legal terms, but also enhances the ability to process complex emotional expressions, making legal consultation more humane and able to respond to user emotional changes in a timely manner, thereby improving user satisfaction and trust.

[0031] 2. The present invention combines sentiment analysis with contextual understanding, and the system can provide users with tailored legal advice. This personalized service fully considers the user's emotional state and historical consultation data, so that the output legal advice not only meets the user's needs in terms of professionalism, but also responds on the emotional level, forming a good interactive experience, thereby enhancing the user's sense of participation and dependence.

[0032] 3. The multimodal interactive functions (such as voice recognition and OCR) and efficient information integration modules integrated in the present invention significantly improve the speed and accuracy of information acquisition and processing. Users can conveniently submit legal issues through a variety of input methods, and the system can quickly extract and analyze relevant information and generate response suggestions in real time. This efficient workflow not only improves the response time of consultation, but also effectively reduces the complexity and cognitive burden of users in the legal consultation process, thereby optimizing the overall user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a system framework diagram of the present invention;

[0034] Figure 2 It is a system diagram of the intermediate layer processing module of the present invention;

[0035] Figure 3 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0036] The following will be combined with the drawings in the specification of the present invention to clearly and completely describe the technical solutions 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 creative work are within the scope of protection of the present invention.

[0037] Embodiment 1:

[0038] Please see attached Figure 1 - Attachment Figure 3 The embodiment of the present invention provides an intelligent interactive system based on legal consultation, including:

[0039] The front-end user interface supports text input, voice input, and image input, and has multi-language switching function;

[0040] The middle layer processing modules, including sentiment analysis module, information integration module, context management module and response generation module, can dynamically adjust the processing strategy;

[0041] The back-end legal knowledge base stores laws and regulations, case analysis and answers to common legal questions, and supports real-time updates, version management and efficient retrieval.

[0042] The sentiment analysis module includes:

[0043] A deep learning model for receiving user input and performing emotion recognition. The model is self-learning and can be continuously optimized based on user feedback;

[0044] Sentiment dictionary, which annotates sentiment words related to the legal field, their sentiment categories and intensities, and supports the expansion of user-defined vocabulary.

[0045] Specifically, deep learning model:

[0046] The model is specifically designed to receive and analyze user input information (including text, voice and images) and perform emotion recognition. By using advanced natural language processing (NLP) technology and deep learning algorithms (such as long short-term memory networks (LSTM) or transformer models (Transformer), the model can effectively capture emotional clues in user input.

[0047] The model has the ability to self-learn and can continuously optimize based on real-time user feedback and historical interaction data. This means that as the system is used, the model will gradually improve its emotion recognition accuracy and adapt to the user's personalized expression and emotional changes.

[0048] In order to improve the effect of sentiment recognition, the model used a large amount of law-related corpus during the training process, including user consultation records, legal texts and case analysis, to ensure its recognition ability in the legal context.

[0049] Sentiment Dictionary:

[0050] The sentiment dictionary is another important component of the sentiment analysis module, which contains sentiment words related to the legal field, their corresponding sentiment categories (such as positive, negative, neutral) and sentiment intensity (such as mild, moderate, strong). The construction of this dictionary relies on expert review and data-driven methods to ensure its professionalism and accuracy.

[0051] The dictionary not only provides the model with basic sentiment recognition capabilities, but also supports the expansion of user-defined vocabulary. Users can add new sentiment vocabulary or adjust the sentiment classification and intensity of existing vocabulary according to specific needs to better reflect their specific legal sentiment expressions.

[0052] The dynamic update mechanism of the sentiment dictionary ensures the flexibility and adaptability of the system, enabling it to respond quickly to emerging terms or trends in the legal field.

[0053] The sentiment recognition model is trained based on the BERT model and uses a special data set in the legal field to fine-tune the model, optimize the recognition ability of legal terms, and integrate multiple sentiment analysis algorithms to improve the comprehensive judgment ability of the model.

[0054] The multimodal user interface has a responsive design that can provide a consistent user interaction experience on different devices, and has adaptive layout and theme switching functions to meet the preferences of different users. Different devices include PCs, smartphones and tablets.

[0055] Specifically, emotion recognition model

[0056] The sentiment recognition model is trained based on the BERT model, making full use of its powerful natural language understanding capabilities. The model is fine-tuned using a dedicated dataset in the legal field to optimize the recognition of legal terms and expressions, ensuring the accuracy of sentiment analysis in the legal context. The specific implementation steps include:

[0057] Dataset construction: Collect legal-related text data, including legal documents, case analysis, user consultation records, etc., to ensure that the data set covers a variety of legal scenarios and terms. These data will be annotated with emotional states to provide a training basis for the model.

[0058] Model fine-tuning: Based on the pre-trained BERT model, fine-tuning is performed using the above-mentioned dedicated dataset. By adjusting the model parameters, it can better adapt to the words and context in the legal field, thereby improving the accuracy of sentiment recognition.

[0059] Sentiment analysis algorithm integration: In addition to BERT-based sentiment recognition, the model also integrates a variety of sentiment analysis algorithms (such as sentiment polarity analysis, sentiment intensity assessment, etc.) to improve comprehensive judgment capabilities. This integration method can combine the advantages of different algorithms to provide more comprehensive sentiment analysis results.

[0060] Continuous learning: The model has the ability to self-learn and can be continuously updated based on user feedback and new data to improve the accuracy and adaptability of emotion recognition.

[0061] Multimodal User Interface

[0062] The multimodal user interface adopts a responsive design to provide users with a consistent interactive experience. The specific design features are as follows:

[0063] Consistent experience: Whether the user is using a PC, smartphone or tablet, the interface can automatically adjust the layout and element size to ensure that the user's interactive experience is consistent on different devices. This is achieved through CSS media queries and flexible grid layout technology to ensure that the content can be perfectly displayed on various screen sizes.

[0064] Adaptive layout: The interface design takes into account the operation methods of different devices (such as touch screen and mouse clicks) and adopts adaptive layout to ensure that users can input, browse and operate conveniently. The system will automatically adjust the position and size of interface elements to ensure optimal usability.

[0065] Theme switching function: Users can choose different themes (such as light mode and dark mode) according to personal preferences or environmental changes to improve user comfort and satisfaction. This function is implemented through simple interface control, allowing users to switch quickly.

[0066] Multimodal interaction support: The interface supports text, voice and image input, providing users with a variety of interaction methods. The system will automatically adapt the response strategy according to the user's input form to improve the overall interaction efficiency and user experience.

[0067] The speech recognition module is combined with a legal terminology optimization algorithm to use an efficient speech recognition engine, and optimizes legal terminology for specific scenarios to improve recognition accuracy, and is capable of processing speech inputs with different accents and speaking speeds.

[0068] The image processing module includes an optical character recognition (OCR) function that can extract text information from legal documents uploaded by users, supports processing of files in different formats such as PDF, JPEG, and PNG, and combines NLP technology to perform legal relevance analysis on the extracted content to improve information processing efficiency. The response generation module generates personalized legal advice based on user emotions and input context, and adjusts the response strategy in real time based on historical consulting data and user feedback to ensure that the generated advice is highly relevant and accurate.

[0069] Specifically, the speech recognition module

[0070] The speech recognition module combines an efficient speech recognition engine and a legal term optimization algorithm to improve the accuracy of speech input in the legal consulting system. The specific technical solutions include:

[0071] Efficient speech recognition engine: Using advanced speech recognition technology (such as Google Speech-to-Text or Microsoft Azure Speech Services), it can quickly and accurately convert user voice input into text. These engines have been trained with a large amount of voice data and have high recognition accuracy.

[0072] Legal terminology optimization: To meet the special needs of the legal field, we develop a scenario-specific legal terminology optimization algorithm that can identify and correct legal terminology during speech recognition to ensure accurate capture of legal-related content. For example, we expand the vocabulary of common legal terms and expressions to improve the recognition rate of specific terms.

[0073] Diversified voice input processing: This module can process voice input with different accents and speaking speeds, using adaptive algorithms to identify multiple accent features and dynamically adjust. By training the model to adapt to different pronunciation styles, the overall recognition ability is improved to ensure that the input of different users can be accurately understood.

[0074] Image processing module

[0075] The image processing module integrates an optical character recognition (OCR) function, and the specific implementation scheme is as follows:

[0076] OCR function: Use OCR technology to extract text information from legal documents uploaded by users. Support multiple file formats such as PDF, JPEG and PNG, ensuring that users can easily upload different types of legal documents.

[0077] Legal text processing: After extracting the text, we combine natural language processing (NLP) technology to analyze the legal relevance of the identified content. This process includes grammatical and semantic analysis of the extracted text to identify legal clauses, key cases and related information, and improve the efficiency and accuracy of information processing.

[0078] Data output and feedback: After extraction and analysis, the system converts relevant information into structured data for subsequent reference in legal consultation. This process can be completed automatically, reducing the user's manual operations and improving overall efficiency.

[0079] Response generation module

[0080] The response generation module aims to provide personalized legal advice based on the user's emotional state and input context. The specific scheme includes:

[0081] Personalized legal advice generation: By integrating the sentiment analysis results and historical consultation data input by the user, the system is able to generate legal advice tailored to the user's specific needs. The advice adjusts the language style and content according to the user's emotional state (such as anxiety, anger) to make it more in line with the user's psychological needs.

[0082] Real-time adjustment of response strategies: The system combines user feedback to update and adjust the recommendation generation strategy in real time to ensure that the information and suggestions provided are always highly relevant and accurate. Through machine learning algorithms, the system can learn user preferences and behaviors and optimize future responses.

[0083] Integration of historical consultation data: By leveraging the user’s historical consultation records, the system can consider past interactions in its recommendations, ensuring that the recommendations provided are more comprehensive and coherent in terms of context.

[0084] Embodiment 2:

[0085] Please see attached Figure 1 - Attachment Figure 3 , an embodiment of the present invention provides an intelligent interaction method based on legal consultation, including:

[0086] Collect user text, voice and image input, and use multimodal fusion technology to uniformly process user data in different input forms;

[0087] Preprocess the collected input, including cleaning and noise removal of text data, and transcribing and standardizing speech data to ensure data quality;

[0088] Use sentiment analysis models to identify user emotions, generate sentiment reports, and combine them with user input to provide real-time feedback on emotional status;

[0089] Integrate user input and sentiment reports to generate response suggestions, and use natural language generation (NLG) technology to output professional and easy-to-understand legal consulting content.

[0090] The emotion recognition step comprises:

[0091] Use sentiment dictionaries to extract sentiment features from user input and combine contextual information to identify changes in user emotional states;

[0092] A variety of sentiment analysis algorithms including sentiment polarity analysis and sentiment intensity assessment are applied to comprehensively analyze user input to improve the accuracy of sentiment recognition. The personalized response generation step includes:

[0093] Generate legal advice content that meets user needs based on user emotions and input context, and optimize response strategies based on user feedback to ensure that the legal advice provided meets user expectations in terms of professionalism and personalization. At the same time, record users' historical consultation data to provide more accurate background information for subsequent interactions.

[0094] Specifically, the emotion recognition step aims to accurately analyze the user's emotional state to enhance the interactivity and personalized service of the legal consulting system. The specific solutions include:

[0095] Sentiment feature extraction: The system first uses the sentiment dictionary to extract sentiment features from the user's input. The dictionary contains sentiment words related to the legal field and their sentiment categories and intensities. By analyzing the input text, the system can identify the basic components of the user's emotions.

[0096] Contextual information integration: When performing sentiment analysis, the system not only considers the user's input content, but also combines contextual information, such as the user's historical consultation records and the topic of the current conversation. This comprehensive analysis method helps to identify changes in the user's emotional state and ensure that the system can accurately understand the user's true emotional needs.

[0097] Application of multiple sentiment analysis algorithms: To improve the accuracy of sentiment recognition, the system applies multiple sentiment analysis algorithms, including:

[0098] Sentiment polarity analysis: used to determine whether the sentiment of user input is positive, negative, or neutral.

[0099] Emotional intensity assessment: Evaluate the intensity of the user's emotions to quantify the depth of emotional expression. Through the combined application of these algorithms, the system can more comprehensively and accurately identify the user's emotional state.

[0100] Personalized response generation steps

[0101] Based on emotion recognition, the personalized response generation step further improves the user experience. The specific solution is as follows:

[0102] Legal advice content generation: Based on the user's emotional state and input contextual information, the system is able to generate legal advice content that meets the user's needs. This process uses natural language generation (NLG) technology to ensure that the output advice is professional and easy to understand. At the same time, the system will adjust the tone and style of the advice based on the user's emotional state to better meet the user's emotional needs.

[0103] User feedback optimizes response strategy: The system optimizes response strategy in real time by recording and analyzing user feedback. After each user interaction with the system, the system will evaluate whether the suggestions provided meet the user's expectations and adjust future response generation mechanisms based on the feedback to improve the overall service quality.

[0104] Historical consultation data records: When generating personalized legal advice, the system will record the user's historical consultation data to provide more accurate background information for subsequent interactions. By analyzing these historical data, the system can identify the user's preferences and habits, and thus provide advice that is more in line with the user's needs in subsequent consultations.

[0105] Embodiment three:

[0106] Front-end user interface

[0107] Functional design

[0108] The front-end user interface is the main entrance for users to interact with the system. The design should take into account the diverse needs of users. The interface supports the following functions:

[0109] Multimodal input: Users can interact with the system through text input, voice input, and image input. For example, users can ask legal questions through voice, or upload legal documents to get relevant advice.

[0110] Multi-language support: The interface has a multi-language switching function, ensuring that non-Chinese users can also use the system, expanding its market applicability.

[0111] Adaptive design: Whether the user is using a PC, smartphone or tablet, the system can automatically adjust the layout and style according to the characteristics of the device to provide a consistent user experience.

[0112] User Experience

[0113] The design of the interface should focus on user experience and ensure that users can get immediate feedback when entering information. Users can be guided to use the system effectively through prompt information, sample questions and help functions. The interface should also provide historical consultation records so that users can check previous legal advice at any time.

[0114] Middle layer processing module

[0115] Module Structure

[0116] The middle layer processing module is the core of the system, which mainly includes sentiment analysis module, information integration module, context management module and response generation module.

[0117] Sentiment analysis module: A deep learning model is used to identify user emotions. The model is trained using legal field-related datasets and combined with sentiment dictionaries to annotate the sentiment attributes of legal terms. The system can identify the user's emotional state (such as anxiety, anger, or satisfaction) in real time based on the user's input and generate a corresponding sentiment report.

[0118] Information integration module: This module is responsible for fusing multimodal input data and extracting user intent and key information. Through context management, the system can analyze the user's historical interaction records, understand their consultation background, and ensure the pertinence and accuracy of subsequent processing.

[0119] Response generation module: Generates personalized legal advice based on the user’s input and emotional state. Uses natural language generation (NLG) technology to transform information in the legal knowledge base into easy-to-understand legal advice, ensuring the professionalism and actionability of the advice.

[0120] Processing Flow

[0121] After the user enters the information, the system processes it through the following process:

[0122] Input recognition: The system recognizes the text, voice or image input by the user and performs necessary conversions (such as voice-to-text, OCR processing, etc.).

[0123] Sentiment Analysis: The sentiment analysis module analyzes user input and generates a sentiment status report.

[0124] Information integration: The information integration module extracts the user's intention and key information and combines it with the emotional state to form contextual information.

[0125] Response generation: The response generation module generates legal advice based on the integrated information and feeds it back to the user.

[0126] Sentiment Scoring Model

[0127] The text input by the user is T, the sentiment dictionary is D, and the context information is C. We define a sentiment score S to represent the sentiment intensity and sentiment polarity of the user input. The formula is as follows:

[0128]

[0129] Variable Description

[0130] n: the number of sentiment words in the sentiment dictionary that match the user input T, w i : represents the weight of sentiment word i, which can be obtained through the sentiment intensity in the sentiment dictionary, e i : The sentiment intensity score of sentiment word i in user input T, usually in the range of [-1,1], where -1 indicates strong negativity and 1 indicates strong positivity. α: Weight coefficient, used to adjust the influence of context information on the sentiment score. f(C): Context information function, returns a sentiment adjustment value based on context C, which can be used to increase or decrease the overall sentiment score. For example, if the context contains the user's past negative emotions, f(C) may be negative.

[0131] Backend legal knowledge base

[0132] Database structure

[0133] The back-end legal knowledge base is the knowledge base of the system, including laws and regulations, case analysis, answers to common legal questions and related literature. The structure of the knowledge base is designed as follows:

[0134] Laws and regulations module: includes the latest national and local laws and regulations, and supports version management to ensure that users obtain the latest information.

[0135] Case analysis module: Integrates classic cases and precedent analysis, provides analysis of the application of legal provisions in specific cases, and helps users better understand the law.

[0136] FAQ module: collects users' common legal questions and answers as a quick reference for consultation.

[0137] Data update and management

[0138] The back-end legal knowledge base needs to support real-time updates to ensure the accuracy and timeliness of the information provided. Legal information websites and official release channels can be crawled regularly to update regulations and case data in a timely manner. In addition, manual review and machine learning algorithms are used to ensure the accuracy and consistency of information.

[0139] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent interactive system based on legal consultation, characterized in that: include: The front-end user interface supports text input, voice input, and image input, and has multi-language switching function; The middle layer processing modules, including sentiment analysis module, information integration module, context management module and response generation module, can dynamically adjust the processing strategy; The back-end legal knowledge base stores laws and regulations, case analysis and answers to common legal questions, and supports real-time updates, version management and efficient retrieval.

2. According to claim 1, the intelligent interactive system based on legal consultation is characterized in that: The sentiment analysis module includes: A deep learning model for receiving user input and performing emotion recognition. The model is self-learning and can be continuously optimized based on user feedback; Sentiment dictionary, which annotates sentiment words related to the legal field, their sentiment categories and intensities, and supports the expansion of user-defined vocabulary.

3. According to claim 1, the intelligent interactive system based on legal consultation is characterized in that: The sentiment recognition model is trained based on the BERT model and uses a special data set in the legal field to fine-tune the model, optimize the recognition ability of legal terms, and integrate multiple sentiment analysis algorithms to improve the comprehensive judgment ability of the model.

4. The intelligent interactive system based on legal consultation according to claim 1 is characterized in that: The multimodal user interface has a responsive design that can provide a consistent user interaction experience on different devices, and has adaptive layout and theme switching functions to meet the preferences of different users. Different devices include PCs, smartphones and tablets.

5. The intelligent interactive system based on legal consultation according to claim 1 is characterized in that: The speech recognition module is combined with a legal terminology optimization algorithm to use an efficient speech recognition engine, and optimizes legal terminology for specific scenarios to improve recognition accuracy, and is capable of processing speech inputs with different accents and speaking speeds.

6. The intelligent interactive system based on legal consultation according to claim 1 is characterized in that: The image processing module includes an optical character recognition (OCR) function that can extract text information from legal documents uploaded by users, supports processing of files in different formats such as PDF, JPEG, and PNG, and combines NLP technology to perform legal relevance analysis on the extracted content to improve information processing efficiency.

7. The intelligent interactive system based on legal consultation according to claim 1 is characterized in that: The response generation module generates personalized legal advice based on user sentiment and input context, and adjusts the response strategy in real time based on historical consultation data and user feedback to ensure that the generated advice is highly relevant and accurate.

8. An intelligent interactive method based on legal consultation, according to the intelligent interactive system based on legal consultation according to claims 1-7, characterized in that: include: Collect user text, voice and image input, and use multimodal fusion technology to uniformly process user data in different input forms; Preprocess the collected input, including cleaning and noise removal of text data, and transcribing and standardizing speech data to ensure data quality; Use sentiment analysis models to identify user emotions, generate sentiment reports, and combine them with user input to provide real-time feedback on emotional status; Integrate user input and sentiment reports to generate response suggestions, and use natural language generation (NLG) technology to output professional and easy-to-understand legal consulting content.

9. The intelligent interactive system and method based on legal consultation according to claim 8, characterized in that: The emotion recognition step comprises: Use sentiment dictionaries to extract sentiment features from user input and combine contextual information to identify changes in user emotional states; A variety of sentiment analysis algorithms including sentiment polarity analysis and sentiment intensity assessment are applied to conduct a comprehensive analysis of user input to improve the accuracy of sentiment recognition.

10. The intelligent interactive system and method based on legal consultation according to claim 8, characterized in that: The personalized response generation step includes: Generate legal advice content that meets user needs based on user emotions and input context, and optimize response strategies based on user feedback to ensure that the legal advice provided meets user expectations in terms of professionalism and personalization. At the same time, record users' historical consultation data to provide more accurate background information for subsequent interactions.

Images