Intelligent question and answer method, device and equipment based on character characters and medium

Abstract

The invention discloses an intelligent question answering method and device based on character characters, equipment and a medium, and belongs to the field of artificial intelligence. The method comprises the steps that information extraction is conducted on user questions, and keywords are obtained; searching data of a corresponding user according to the keyword to obtain associated data; wherein the associated data comprises text associated data, image associated data and video associated data; performing multi-source data feature extraction based on the type of the associated data to obtain a first character character feature; performing MBT I character analysis according to the first character character characteristics to obtain second character character characteristics; and generating an answer according to the user question, the keyword, the first character character feature and the second character character feature, and outputting an answer text of the user question. Therefore, by implementing the method and the device, accurate answers more fitting the personality of the user can be generated.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a method, device, equipment and medium for intelligent question answering based on character traits. Background Art

[0002] With the continuous development of artificial intelligence technology, intelligent question-answering systems have gained widespread application. As a key branch of this field, intelligent question-answering systems based on personality analysis combine psychology, behavioral science, and natural language processing techniques to analyze the personality traits displayed by users during conversations, better understand user intent, and provide accurate responses that are more tailored to their personalities. This system not only enhances the user experience but also plays a vital role in multiple application scenarios, including personalized recommendations, emotional computing, and intelligent customer service.

[0003] Currently, intelligent question-answering systems based on personality analysis are not fully mature, but several related technologies have been applied in this field. Affective computing: Some intelligent question-answering systems use sentiment analysis to infer users' emotions and select appropriate response styles. Personalized recommendation systems: Based on users' historical behavior or expressed preferences, personalized recommendation systems use machine learning methods to provide users with tailored information. Research integrating psychology with natural language processing: Some studies have begun to combine psychological theories with natural language processing techniques for personality analysis. These studies provide references for personality analysis in intelligent question-answering systems.

[0004] Despite some technological advances, current intelligent question-answering systems based on personality analysis still face numerous drawbacks and challenges. First, the accuracy of personality analysis is insufficient: Existing personality analysis methods mostly rely on textual data, and the information in text often does not fully reflect a person's overall personality. For example, the emotions and attitudes expressed by users in certain situations may deviate from their long-term personality traits. Second, the ability to personalize responses is insufficient: While some intelligent question-answering systems can generate responses based on user input, the level of personalization required for users with different personalities is insufficient. They fail to fully consider the user's personality traits, preferences, and emotional state to generate responses that are more tailored to the user's personality, thus impacting the user experience. Finally, there are data privacy and ethical issues: To effectively analyze personality, intelligent question-answering systems require the collection of large amounts of user data, particularly content related to emotions and attitudes. This raises privacy and ethical concerns, as users may be reluctant to share this sensitive information without explicit authorization. Summary of the Invention

[0005] The present invention provides a method, device, equipment and medium for intelligent question answering based on character, which can analyze character through multi-source data and generate more accurate answers based on the comprehensive analysis results.

[0006] An embodiment of the present invention provides an intelligent question-answering method based on character traits, comprising:

[0007] Extract information from user questions and obtain keywords;

[0008] Searching for information about the corresponding user based on the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data, and video associated data;

[0009] Perform multi-source data feature extraction based on the type of the associated data to obtain a first character personality feature;

[0010] Performing MBTI personality analysis based on the first character's personality traits to obtain the second character's personality traits;

[0011] An answer is generated according to the user question, keywords, the first character's personality traits and the second character's personality traits, and an answer text of the user question is output.

[0012] By understanding user questions, the embodiments of the present invention can better extract user information, which is helpful for the subsequent acquisition of comprehensive user-related data; by searching for the corresponding user's information through keywords, it can fully obtain the user's related data, which is helpful for the subsequent accurate personality analysis of the user; by extracting features through multi-source data, it can comprehensively analyze the user's personality, which is helpful for the subsequent generation of more accurate answers; by performing MBTI personality analysis based on the personality analysis results, it can further enrich the description of the personality characteristics of the person, which is helpful for the subsequent generation of more accurate answers; by generating answers by integrating user questions, keywords and personality characteristics, it can generate accurate answers that are more suitable for the user's personality. Compared with the existing technology that uses single text data for personality analysis and question and answer generation, the present application can use multi-source data to perform personality analysis and generate more accurate answers based on the comprehensive analysis results.

[0013] Furthermore, the information extraction from the user question to obtain keywords includes:

[0014] Perform semantic analysis on the user question using a large language model;

[0015] According to the semantic analysis result, keywords related to the user in the user question are extracted.

[0016] The embodiment of the present invention performs semantic analysis on user questions through a large language model, which can better extract user information and facilitate the subsequent acquisition of comprehensive user-related data.

[0017] Furthermore, searching for the corresponding user's information based on the keyword to obtain the associated data includes:

[0018] Searching for information about a corresponding person in a preset person knowledge base based on the keyword;

[0019] If the information of the corresponding person does not exist in the person knowledge base, searching for the information of the corresponding person in relevant web pages according to the keyword;

[0020] The found information is subjected to data standardization to obtain related data.

[0021] The embodiment of the present invention searches for the information of the corresponding user through keywords, and can fully obtain the user's related data, which is helpful for subsequent accurate character analysis of the user.

[0022] Furthermore, the data found is subjected to data standardization processing to obtain related data, including:

[0023] If the data is text, the data is cleaned and formatted to obtain text-related data;

[0024] If the data is an image, performing image processing and feature extraction on the data to obtain image-related data;

[0025] If the material is a video, key frames in the material are extracted, and image processing and feature extraction are performed on the key frames to obtain first video-related data; speech recognition is performed on the material, and the recognized speech is used as second video-related data; the speech is converted into text, and data cleaning and formatting are performed on the text to obtain third video-related data.

[0026] The embodiment of the present invention provides standardized input data for subsequent character analysis by performing standardization processing on multi-source data.

[0027] Furthermore, the multi-source data feature extraction based on the type of the associated data to obtain the first character personality feature includes:

[0028] If the associated data is text-related data, extracting features from the text-related data using a large language model to obtain text-based character traits;

[0029] If the associated data is image associated data, then feature extraction is performed on the image associated data using a multimodal large model to obtain image-based character features of the person;

[0030] If the associated data is video-associated data, feature extraction is performed on the first video-associated data using the multimodal large model to obtain a first video-based character personality feature; feature extraction is performed on the second video-associated data using the voice large model to obtain a second video-based character personality feature; and feature extraction is performed on the third video-associated data using the language large model to obtain a third video-based character personality feature.

[0031] The text-based character traits, the image-based character traits, or the video-based character traits are combined to obtain a first character trait.

[0032] The embodiment of the present invention extracts features from multi-source data, which can comprehensively analyze the user's personality and help generate more accurate answers later.

[0033] Furthermore, performing MBTI personality analysis based on the first character's personality traits to obtain the second character's personality traits includes:

[0034] Perform feature mapping on the personality traits of the first character through the language big model to obtain the MBTI type of the corresponding character;

[0035] According to the MBTI type, a corresponding personality trait description is searched in a preset personality knowledge base to obtain the personality characteristics of the second person.

[0036] The embodiment of the present invention performs MBTI personality analysis based on the personality analysis results, which can further enrich the description of personality characteristics and help generate more accurate answers later.

[0037] Furthermore, generating an answer based on the user question, keywords, the first character's personality traits, and the second character's personality traits, and outputting an answer text to the user question, includes:

[0038] Combining the user question, the keyword, the first character's personality trait, and the second character's personality trait to obtain a character's personality trait text;

[0039] Generate an answer based on the character's personality traits text using a large language model, and output the answer text to the user's question.

[0040] The embodiment of the present invention generates answers by integrating user questions, keywords and character traits, and can generate accurate answers that are more in line with the user's personality.

[0041] Another embodiment of the present invention further provides an intelligent question-answering device based on character, comprising: a keyword module, a related data module, a first character character feature module, a second character character feature module and a question answering module;

[0042] The keyword module is used to extract information from user questions and obtain keywords;

[0043] The associated data module is used to search for information of the corresponding user according to the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data and video associated data;

[0044] The first character feature module is configured to extract features from multi-source data based on the type of the associated data to obtain the first character feature;

[0045] The second character trait module is configured to perform MBTI personality analysis based on the first character trait to obtain a second character trait;

[0046] The question answering module is used to generate an answer based on the user question, keywords, the first character's personality traits and the second character's personality traits, and output an answer text for the user question.

[0047] Another embodiment of the present invention also provides a terminal device, including: a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the steps of an intelligent question-answering method based on character personality as described in the present invention.

[0048] Another embodiment of the present invention further provides a computer-readable storage medium item, comprising: a stored computer program, which, when the computer program is running, controls the device where the computer-readable storage medium is located to execute the steps of an intelligent question-answering method based on character personality of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 A schematic diagram of a process flow of a character-based intelligent question-answering method provided by an embodiment of the present invention;

[0050] Figure 2 A schematic diagram of the process of the character-based intelligent question-answering system provided by an embodiment of the present invention;

[0051] Figure 3 A schematic diagram of the structure of a character-based intelligent question-answering device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] To make the objectives, technical solutions, and advantages of this application more clear, the technical solutions in this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative efforts are within the scope of protection of this application.

[0053] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned figure descriptions are intended to cover non-exclusive inclusions.

[0054] In the description of the embodiments of this application, the technical terms "first" and "second" are used only to distinguish different objects and should not be understood to indicate or imply relative importance or implicitly specify the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise clearly and specifically defined.

[0055] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0056] In the description of the embodiments of this application, the term "and / or" is simply a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent the following three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this document generally indicates that the associated objects are in an "or" relationship.

[0057] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0058] See also Figure 1 To solve the problem in the prior art that accurate answers that match the character's personality cannot be obtained by generating questions and answers based on single text data, an embodiment of the present invention provides an intelligent question-answering method based on character, comprising steps S101 to S105;

[0059] Step S101: extract information from user questions to obtain keywords.

[0060] Furthermore, step S101 includes:

[0061] Perform semantic analysis on the user question using a large language model;

[0062] According to the semantic analysis result, keywords related to the user in the user question are extracted.

[0063] Preferably, the DeepSeek-R1 model is used to perform semantic analysis on user questions, extract user-related keywords from user questions and form a question summary.

[0064] The embodiment of the present invention performs semantic analysis on user questions through a large language model, which can better extract user information and facilitate the subsequent acquisition of comprehensive user-related data.

[0065] Step S102 , searching for information of the corresponding user according to the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data and video associated data.

[0066] Furthermore, step S102 includes:

[0067] Searching for information about a corresponding person in a preset person knowledge base based on the keyword;

[0068] If the information of the corresponding person does not exist in the person knowledge base, searching for the information of the corresponding person in relevant web pages according to the keyword;

[0069] The found information is subjected to data standardization to obtain related data.

[0070] Preferably, the preset character knowledge base is a personal information knowledge set consisting of personal text data, image data and video data; one user corresponds to one character knowledge base, which can ensure the independence of user data and avoid the risk of personal information leakage caused by using a public database.

[0071] Preferably, Python web crawler technology is used to search for information about corresponding people in relevant web pages; the crawler constructs a search strategy based on keywords, such as searching for relevant web pages and social media information of a specific person in a search engine; during the crawling process, the website's robots.txt rules are followed to avoid placing too much burden on the target website; for text data, the crawler extracts the main text content from the web page and removes irrelevant elements such as advertisements and navigation bars; for image data, the crawler crawls according to the image links and related tags in the web page, and records the source and context information of the image; for video data, after crawling the video link, the video is downloaded through a third-party tool, and the video metadata, such as the title and release time, is extracted.

[0072] The embodiment of the present invention searches for the information of the corresponding user through keywords, and can fully obtain the user's related data, which is helpful for subsequent accurate character analysis of the user.

[0073] Furthermore, the data found is subjected to data standardization processing to obtain related data, including:

[0074] If the data is text, the data is cleaned and formatted to obtain text-related data;

[0075] If the data is an image, performing image processing and feature extraction on the data to obtain image-related data;

[0076] If the material is a video, key frames in the material are extracted, and image processing and feature extraction are performed on the key frames to obtain first video-related data; speech recognition is performed on the material, and the recognized speech is used as second video-related data; the speech is converted into text, and data cleaning and formatting are performed on the text to obtain third video-related data.

[0077] Preferably, if the data is text, natural language processing technology is used for cleaning, such as removing stop words, correcting spelling errors and performing stem extraction, and the text is converted into vector representation or structured data according to the data format requirements; if the data is an image, an image processing library is used to perform denoising, adjust brightness and contrast, and zoom, etc., so that it meets the input requirements of the large model, and the image data is further converted into feature vectors, such as extracting the features of the image data through a convolutional neural network; if the data is a video, a video processing tool is used to extract key frames, and the key frames are processed similarly to the image data. At the same time, speech recognition is performed on the voice in the video, and the voice is converted into text data, and then the text data is cleaned and formatted.

[0078] The embodiment of the present invention provides standardized input data for subsequent character analysis by performing standardization processing on multi-source data.

[0079] Step S103 : extracting features from multi-source data based on the type of the associated data to obtain a first character characteristic.

[0080] Furthermore, step S103 includes:

[0081] If the associated data is text-related data, extracting features from the text-related data using a large language model to obtain text-based character traits;

[0082] If the associated data is image associated data, then feature extraction is performed on the image associated data using a multimodal large model to obtain image-based character features of the person;

[0083] If the associated data is video-associated data, feature extraction is performed on the first video-associated data using the multimodal large model to obtain a first video-based character personality feature; feature extraction is performed on the second video-associated data using the voice large model to obtain a second video-based character personality feature; and feature extraction is performed on the third video-associated data using the language large model to obtain a third video-based character personality feature.

[0084] The text-based character traits, the image-based character traits, or the video-based character traits are combined to obtain a first character trait.

[0085] Preferably, if the associated data is text-associated data, the DeepSeek-R1 model is used to analyze the keywords, emotional vocabulary, and sentences describing the character's behavior and psychology in the text-associated data; if the associated data is image-associated data, the QVQ-Max model is used to analyze the character's facial features, body posture, and background environment in the image-associated data; if the associated data is video-associated data, the QVQ-Max model is first used to analyze the character's appearance and expressions (such as smiles, frowns, and changes in eye contact), body movements (gesture amplitude, walking posture, and stretching or restraint of body posture, etc.), and behavior patterns (whether actively participating in activities or being relatively passive, etc.) in the first video-associated data; secondly, the Qwen2_Audio model is used to analyze the speaking speed, volume, and intonation in the second video-associated data; finally, the DeepSeek-R1 model is used to perform text data analysis on the third video-associated data.

[0086] Exemplarily, the first character's personality traits are text data. For example, the personality traits of a person who frequently shows a cheerful smile or has large and relaxed body movements are described as extroverted and lively; the personality traits of a person who always frowns or has relatively restrained body movements are described as introverted and cautious; the personality traits of a person who speaks fast and with varied tones are described as enthusiastic and quick-thinking; and the personality traits of a person who speaks softly and with a flat tone are described as gentle and steady.

[0087] The embodiment of the present invention extracts features from multi-source data, which can comprehensively analyze the user's personality and help generate more accurate answers later.

[0088] Step S104: performing MBTI personality analysis based on the first character's personality traits to obtain a second character's personality traits.

[0089] Furthermore, step S104 includes:

[0090] Perform feature mapping on the personality traits of the first character through the language big model to obtain the MBTI type of the corresponding character;

[0091] According to the MBTI type, a corresponding personality trait description is searched in a preset personality knowledge base to obtain the personality characteristics of the second person.

[0092] Preferably, the DeepSeek-R1 model is used to perform feature mapping on the personality traits of the first person to obtain the MBTI type of the corresponding person.

[0093] Exemplarily, the second character traits are text data, for example, describing the character traits of a person whose MBTI type is ENTJ as self-confidence, decisiveness, and leadership.

[0094] The embodiment of the present invention performs MBTI personality analysis based on the personality analysis results, which can further enrich the description of personality characteristics and help generate more accurate answers later.

[0095] Step S105 , generating an answer based on the user question, keywords, the first character's personality traits, and the second character's personality traits, and outputting an answer text to the user question.

[0096] Furthermore, step S105 includes:

[0097] Combining the user question, the keyword, the first character's personality trait, and the second character's personality trait to obtain a character's personality trait text;

[0098] Generate an answer based on the character's personality traits text using a large language model, and output the answer text to the user's question.

[0099] Preferably, the DeepSeek-R1 model is used to generate answers based on the character personality feature text, and output an answer text with character profile results.

[0100] The embodiment of the present invention generates answers by integrating user questions, keywords and character traits, and can generate accurate answers that are more in line with the user's personality.

[0101] like Figure 2 As shown, based on the above method embodiment, a corresponding system embodiment is provided;

[0102] Use a large language model to perform semantic analysis on user questions, extract user-related keywords from user questions, and form a question summary;

[0103] Search for the corresponding user's information based on the keyword. If there is no relevant information in the character knowledge base, crawl the relevant information in real time using Python web crawler technology;

[0104] Performing data cleaning and formatting operations on the relevant information to obtain standardized data;

[0105] If the standardized data is text data, a language large model is used to perform character analysis; if the standardized data is image data, a multimodal large model is used to perform character analysis; if the standardized data is video data, a multimodal large model, a speech large model, and a language large model are used to perform character analysis; the character analysis results are integrated into a first character character trait description text;

[0106] Based on the first character's personality characteristics description text, perform MBTI personality analysis using the language model, and map the MBTI personality analysis results into the second character's personality characteristics description text;

[0107] The user question, keywords, the first character's personality description text and the second character's personality description text are used as input data for the language model, and the answer text with the character profile results is output.

[0108] like Figure 3 As shown, based on the above method embodiment, a corresponding device embodiment is provided;

[0109] An embodiment of the present invention provides an intelligent question-answering device based on character, comprising: a keyword module 301, a related data module 302, a first character character feature module 303, a second character character feature module 304, and a question answering module 305;

[0110] The keyword module 301 is used to extract information from user questions and obtain keywords;

[0111] The associated data module 302 is used to search for the corresponding user's information according to the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data and video associated data;

[0112] The first character feature module 303 is configured to extract features from multi-source data based on the type of the associated data to obtain a first character feature;

[0113] The second character trait module 304 is configured to perform MBTI personality analysis based on the first character trait to obtain a second character trait;

[0114] The question answering module 305 is used to generate an answer based on the user question, keywords, the first character's personality traits and the second character's personality traits, and output an answer text to the user question.

[0115] In the embodiment of the present invention, the keyword module 301 includes: a semantic analysis submodule and a keyword extraction submodule;

[0116] The semantic analysis submodule is used to perform semantic analysis on the user question using a large language model;

[0117] The keyword extraction submodule is used to extract user-related keywords from the user question based on the semantic analysis result.

[0118] The embodiment of the present invention performs semantic analysis on user questions through a large language model, which can better extract user information and facilitate the subsequent acquisition of comprehensive user-related data.

[0119] In the embodiment of the present invention, the associated data module 302 includes: a character knowledge base submodule, a webpage query submodule and a data processing submodule;

[0120] The character knowledge base submodule is used to search for information about corresponding characters in a preset character knowledge base according to the keywords;

[0121] The webpage query submodule is used to search for the information of the corresponding person in relevant webpages according to the keyword if the information of the corresponding person does not exist in the person knowledge base;

[0122] The data processing submodule is used to perform data standardization processing on the found information to obtain related data.

[0123] The embodiment of the present invention searches for the information of the corresponding user through keywords, and can fully obtain the user's related data, which is helpful for subsequent accurate character analysis of the user.

[0124] In an embodiment of the present invention, the data processing submodule includes: a text processing unit, an image processing unit, and a video processing unit;

[0125] The text processing unit is configured to clean and format the data to obtain text-related data if the data is text;

[0126] The image processing unit is configured to perform image processing and feature extraction on the data to obtain image-related data if the data is an image;

[0127] The video processing unit is used to extract key frames from the material if the material is a video, and perform image processing and feature extraction on the key frames to obtain first video-related data; perform speech recognition on the material and use the recognized speech as second video-related data; convert the speech into text, and perform data cleaning and formatting on the text to obtain third video-related data.

[0128] The embodiment of the present invention provides standardized input data for subsequent character analysis by performing standardization processing on multi-source data.

[0129] In the embodiment of the present invention, the first character feature module 303 includes: a text data submodule, an image data submodule, a video data submodule and a feature synthesis submodule;

[0130] The text data submodule is configured to extract features from the text-related data using a large language model to obtain text-based character traits if the related data is text-related data;

[0131] The image data submodule is configured to, if the associated data is image associated data, extract features from the image associated data using a multimodal large model to obtain image-based character features;

[0132] The video data submodule is configured to, if the associated data is video-associated data, perform feature extraction on the first video-associated data using a multimodal large model to obtain a first video-based character personality feature; perform feature extraction on the second video-associated data using a voice large model to obtain a second video-based character personality feature; and perform feature extraction on the third video-associated data using a language large model to obtain a third video-based character personality feature;

[0133] The feature synthesis submodule is used to combine the text-based character personality feature, the image-based character personality feature or the video-based character personality feature to obtain a first character personality feature.

[0134] The embodiment of the present invention extracts features from multi-source data, which can comprehensively analyze the user's personality and help generate more accurate answers later.

[0135] In the embodiment of the present invention, the second character feature module 304 includes: a feature mapping submodule and a character knowledge base submodule;

[0136] The feature mapping submodule is used to perform feature mapping on the personality traits of the first person through the language model to obtain the MBTI type of the corresponding person;

[0137] The personality knowledge base submodule is used to search for corresponding personality characteristics descriptions in a preset personality knowledge base according to the MBTI type to obtain the personality characteristics of the second person.

[0138] The embodiment of the present invention performs MBTI personality analysis based on the personality analysis results, which can further enrich the description of personality characteristics and help generate more accurate answers later.

[0139] In the embodiment of the present invention, the question answering module 305 includes: a data combining submodule and an answer generating submodule;

[0140] The data combining submodule is used to combine the user question, keywords, the first character's personality traits and the second character's personality traits to obtain a character's personality traits text;

[0141] The answer generation submodule is used to generate answers based on the character personality feature text through a language model and output the answer text of the user question.

[0142] The embodiment of the present invention generates answers by integrating user questions, keywords and character traits, and can generate accurate answers that are more in line with the user's personality.

[0143] It can be understood that the above-mentioned device embodiment corresponds to the method embodiment of the present invention, which can implement an intelligent question-answering method based on character personality provided by any of the above-mentioned method embodiments of the present invention.

[0144] The embodiment of the present invention uses a keyword module to understand user questions, which can better extract user information and help to obtain comprehensive user-related data in the future. The related data module searches for the corresponding user's information based on keywords, which can fully obtain the user's related data and help to perform accurate personality analysis on the user in the future. The first personality feature module extracts features based on multi-source data, which can comprehensively analyze the user's personality and help to generate more accurate answers in the future. The second personality feature module performs MBTI personality analysis based on the personality analysis results, which can further enrich the personality feature description and help to generate more accurate answers in the future. The question answering module generates answers by combining user questions, keywords, and personality features, which can generate accurate answers that are more suitable for the user's personality. Compared with the existing technology that uses single text data to perform personality analysis and question and answer generation, the present application can use multi-source data to perform personality analysis and generate more accurate answers based on the comprehensive analysis results.

[0145] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. Furthermore, in the drawings of the device embodiments provided by the present invention, the connection relationship between modules indicates that they have a communication connection, which may be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement the present invention without inventive effort.

[0146] Based on the above-mentioned embodiment of an intelligent question-answering method based on character, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements an intelligent question-answering method based on character according to any embodiment of the present invention.

[0147] For example, in this embodiment, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more module elements may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0148] The terminal device may be a computing device such as a desktop computer, a notebook computer, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0149] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, connecting various parts of the entire terminal device using various interfaces and lines.

[0150] Based on the above-mentioned method embodiments, another embodiment of the present invention provides a computer-readable storage medium, including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute an intelligent question-answering method based on character personality as described in any one of the above-mentioned method embodiments of the present invention.

[0151] Wherein, the module / unit integrated in the device / terminal equipment, if implemented in the form of a software functional unit and sold or used as an independent product, can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the process in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0152] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An intelligent question-answering method based on character, characterized in that: include: Extract information from user questions and obtain keywords; Searching for information about the corresponding user based on the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data, and video associated data; Perform multi-source data feature extraction based on the type of the associated data to obtain a first character personality feature; Performing MBTI personality analysis based on the first character's personality traits to obtain the second character's personality traits; An answer is generated according to the user question, keywords, the first character's personality traits and the second character's personality traits, and an answer text of the user question is output.

2. The intelligent question-answering method based on character personality according to claim 1, wherein: The information extraction from the user question is performed to obtain keywords, including: Perform semantic analysis on the user question using a large language model; According to the semantic analysis result, keywords related to the user in the user question are extracted.

3. The intelligent question-answering method based on character personality according to claim 1, wherein: The step of searching for the corresponding user's information based on the keyword to obtain the associated data includes: Searching for information about a corresponding person in a preset person knowledge base based on the keyword; If the information of the corresponding person does not exist in the person knowledge base, searching for the information of the corresponding person in relevant web pages according to the keyword; The found information is subjected to data standardization to obtain related data.

4. The intelligent question-answering method based on character personality according to claim 3, characterized in that: The data found is subjected to data standardization processing to obtain related data, including: If the data is text, the data is cleaned and formatted to obtain text-related data; If the data is an image, performing image processing and feature extraction on the data to obtain image-related data; If the material is a video, key frames in the material are extracted, and image processing and feature extraction are performed on the key frames to obtain first video-related data; speech recognition is performed on the material, and the recognized speech is used as second video-related data; the speech is converted into text, and data cleaning and formatting are performed on the text to obtain third video-related data.

5. The intelligent question-answering method based on character personality according to claim 4, characterized in that: The extracting features from multi-source data based on the type of the associated data to obtain the first character feature includes: If the associated data is text-related data, extracting features from the text-related data using a large language model to obtain text-based character traits; If the associated data is image associated data, then feature extraction is performed on the image associated data using a multimodal large model to obtain image-based character features of the person; If the associated data is video-associated data, feature extraction is performed on the first video-associated data using the multimodal large model to obtain a first video-based character personality feature; feature extraction is performed on the second video-associated data using the voice large model to obtain a second video-based character personality feature; and feature extraction is performed on the third video-associated data using the language large model to obtain a third video-based character personality feature. The text-based character traits, the image-based character traits, or the video-based character traits are combined to obtain a first character trait.

6. The intelligent question-answering method based on character traits according to claim 1, wherein: The performing of MBTI personality analysis based on the first character's personality traits to obtain the second character's personality traits includes: Perform feature mapping on the personality traits of the first character through the language big model to obtain the MBTI type of the corresponding character; According to the MBTI type, a corresponding personality trait description is searched in a preset personality knowledge base to obtain the personality characteristics of the second person.

7. The intelligent question-answering method based on character personality according to claim 1, wherein: Generating an answer based on the user question, keywords, the first character's personality traits, and the second character's personality traits, and outputting an answer text to the user question, includes: Combining the user question, the keyword, the first character's personality trait, and the second character's personality trait to obtain a character's personality trait text; Generate an answer based on the character's personality traits text using a large language model, and output the answer text to the user's question.

8. An intelligent question-answering device based on character, characterized in that: include: Keyword module, related data module, first character personality trait module, second character personality trait module and question answering module; The keyword module is used to extract information from user questions and obtain keywords; The associated data module is used to search for information of the corresponding user according to the keyword to obtain associated data; wherein the associated data includes text associated data, image associated data and video associated data; The first character feature module is configured to extract features from multi-source data based on the type of the associated data to obtain the first character feature; The second character trait module is configured to perform MBTI personality analysis based on the first character trait to obtain a second character trait; The question answering module is used to generate an answer based on the user question, keywords, the first character's personality traits and the second character's personality traits, and output an answer text for the user question.

9. A terminal device, characterized in that: The invention comprises a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, an intelligent question-answering method based on character personality as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that include: A stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute an intelligent question-answering method based on character personality as described in any one of claims 1 to 7.

Images