User question and answer method and device, electronic equipment, storage medium and program product

Abstract

The invention provides a user question and answer method and device, electronic equipment, a storage medium and a program product, relates to the technical field of artificial intelligence, and aims to realize picture and text combined answer by searching character analysis and picture display through an equipment ID, a user question and a text answer template. Aiming at the field of question and answer application of equipment such as household appliances, the invention provides a fool-type method for guiding a user to use or operate the equipment, and also provides an applicable man-machine interaction process. According to the method, the text answer template is matched according to the equipment ID, so that the preliminary answer of the user question is quickly obtained. The answering scheme containing character analysis and display pictures is obtained through further searching of the device ID and the user questions, so that the answering scheme is more vivid, improvement of user understanding is facilitated, and the accuracy and efficiency of user question answering are improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a user question and answer method, device, electronic device, storage medium and program product. Background Art

[0002] Existing user questions and answers for household appliances mainly include manual questions and answers, which are essentially a special type of document questions and answers or knowledge questions and answers.

[0003] However, in the field of Q&A on manuals of home appliances and other equipment, existing document Q&A or knowledge Q&A has the following shortcomings: There are a large number of pictures in the manuals of home appliances and other equipment to assist users in operation. The text and pictures constitute a relatively complete information set, while general knowledge Q&A is usually presented to users in the form of text.

[0004] Whether it is manual Q&A or document knowledge Q&A, the common shortcomings of the solutions are as follows: the answers given only contain text, which cannot meet the user's need for guidance on "foolproof" use of home appliances and other equipment. For user question retrieval, it is necessary to obtain multiple (K) candidate knowledge and combine them with a large language model for knowledge Q&A. Due to the contradiction between the K value and the retrieval efficiency, it is difficult to achieve a balance between a good Q&A effect and a low-latency interactive experience, which affects the user experience. When the manual or Q&A template corresponding to the device model does not exist, there is a situation where the user's question cannot be answered or the answer quality is not high. Therefore, the efficiency and accuracy of existing user Q&A for home appliances are low. Summary of the invention

[0005] The present application provides a user question and answer method, device, electronic device, storage medium and program product, which are used to solve the defects of low efficiency and accuracy of manual question and answer of household appliances in the prior art, and to improve the efficiency and accuracy of user question and answer of household appliances.

[0006] In a first aspect, the present application provides a user question and answer method, including: obtaining a user question and a device ID corresponding to the user question; when the device ID exists in a standard question database, matching the user question with the standard question database to determine the standard question, and determining a text answer template corresponding to the user question through the standard question; classifying the user question to determine the type of the user question; when it is determined that the question type is a type of question that needs to be answered in combination with a picture, searching based on the device ID, user question and text answer template, obtaining text analysis and displaying pictures to determine a solution.

[0007] In one embodiment, the user question and answer method also includes: when the device ID does not exist in the standard question database, obtaining device information corresponding to the device ID, searching the device's manual content over the Internet based on the device information; generating a first prompt instruction based on the device ID and the user's question, and obtaining text analysis from the device's manual content based on the first prompt instruction to determine a solution.

[0008] In one embodiment, the user question and answer method also includes: when it is determined that the question type is not a question type that needs to be answered in conjunction with a picture, matching with the device manual database based on the device ID to obtain the device manual content; generating a first prompt instruction based on the device ID and the user question, and obtaining a text analysis from the device manual content based on the first prompt instruction to determine a solution.

[0009] In one embodiment, a search is performed based on the device ID, user question and text answer template, and a text analysis and display image are obtained to determine a solution, including: matching the device ID with the image database to determine the image information corresponding to the device ID, the image information including the image ID and the image description; generating a second prompt based on the image information, user question and text answer template; rearranging the image information and user question based on the second prompt to obtain a text analysis containing the image ID; obtaining a display image from the image database based on the image ID; and obtaining a solution based on the display image and the text analysis.

[0010] In one embodiment, a search is performed based on the device ID, user question, and text answer template, and a text analysis and display image are obtained to determine a solution, including: obtaining an initial display image of the device input by the user; searching the device's manual content based on the device ID, and using at least one of the device's manual content and the text answer template as text data; fusing and organizing the user question, the initial display image, and the text data, and obtaining a text analysis and display image to determine a solution.

[0011] In one embodiment, user questions are matched with a standard question database to determine standard questions, including: converting the user question into a user question text vector, obtaining the similarity between the user question text vector and each standard question text vector in the standard question database, and taking the question corresponding to the standard question text vector with the greatest similarity as the standard question.

[0012] In the second aspect, the present application also provides a user question and answer device, including: a device identification module, used to obtain user questions and a device ID corresponding to the user questions; a question and answer template matching module, used to match the user questions with the standard question database to determine the standard questions when the device ID exists in the standard question database, and determine the text answer template corresponding to the user question through the standard questions; a question type determination module, used to classify user questions to determine the question type of the user questions; a picture and text question and answer module, used to search according to the device ID, user questions and text answer templates when it is determined that the question type is a question and answer type that needs to be answered in combination with pictures, obtain text analysis and display pictures to determine a solution.

[0013] In a third aspect, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, any one of the user question and answer methods described above is implemented.

[0014] In a fourth aspect, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the user question-and-answer methods described above.

[0015] In a fifth aspect, the present application also provides a computer program product, including a computer program, which implements any of the above-mentioned user question and answer methods when executed by a processor.

[0016] The user question and answer method, device, electronic device, storage medium and program product provided by this application realizes a picture-based answer by searching text analysis and displaying pictures through device ID, user questions and text answer templates. This application provides a "fool-proof" method for guiding users to use or operate a device in the field of Q&A applications for home appliances and other devices, and proposes an applicable human-computer interaction process. This application matches the text answer template according to the device ID to quickly obtain preliminary answers to user questions. Further searching through the device ID and user questions to obtain solutions containing text analysis and display pictures makes the solution more vivid, which is conducive to improving user understanding and improving the accuracy and efficiency of user questions and answers. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1This is one of the flow charts of the user question and answer method provided by this application.

[0019] Figure 2 Schematic diagram of the standard question database provided by this application.

[0020] Figure 3 It is a schematic diagram of the question-answer template database provided by this application.

[0021] Figure 4 It is a schematic diagram of the equipment specification database provided in this application.

[0022] Figure 5 It is a schematic diagram of the image database provided by this application.

[0023] Figure 6 This is the second flow chart of the user question and answer method provided by this application.

[0024] Figure 7 This is the third flow chart of the user question and answer method provided by this application.

[0025] Figure 8 It is a structural schematic diagram of the user question and answer device provided by the present invention.

[0026] Fig. 9 It is a structural schematic diagram of the electronic device provided by this application. DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are 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 creative work are within the scope of protection of this application.

[0028] Combine the following Figure 1-Figure 9 Describe the user question and answer method, device and electronic device of the present application.

[0029] Figure 1 This is one of the flow charts of the user question and answer method provided by this application, such as Figure 1 As shown, the user question and answer method includes S100 to S400, and each step is specifically as follows.

[0030] S100: Obtain a user question and a device ID corresponding to the user question.

[0031] like Figure 6As shown, the executor of the embodiment of the present application is a user question and answer device, which includes a device identification module, a question and answer template matching module, a question type determination module and a picture and text question and answer module.

[0032] For questions and answers about the manuals of home appliances and other devices, users are usually required to enter information such as the device model, such as the device ID. However, due to the hidden location of the label on the device, the label is missing, etc., it is not easy to find the corresponding device ID. Therefore, the device identification module can optionally identify the device ID based on the device picture taken by the user. Furthermore, the device category can also be identified through the device picture.

[0033] For example, the device identification module includes a device identification model, whose input is a device image and whose output is a device ID and device category (such as rice cooker, air fryer, etc.).

[0034] The device identification model can be constructed based on the following steps.

[0035] (1) Obtain sample device images: Collect sample device images corresponding to each sample device ID, and clean and filter the sample device images to remove low-quality images such as blurred devices and low device proportions. In order to improve the classification effect of the device recognition model for device images of different qualities, it is necessary to perform data enhancement preprocessing on the sample device images to obtain preprocessed sample device images. Data enhancement preprocessing at least includes adding background noise, rotation, or spatial distortion.

[0036] (2) Construct a sample dataset corresponding to the sample device ID, sample device category, and preprocessed sample device images.

[0037] (3) Based on the sample data set constructed above, the preset device image classification model is trained. The preset image classification model includes EfficientNetV2, Swin Transformer, etc. The preset device image classification model is trained to identify the mapping relationship between the sample device ID, sample device category, and sample device image. The training is terminated until the preset device image classification model can accurately output the corresponding device ID and device category based on the sample device image, and the device recognition model is obtained.

[0038] S200: When the device ID exists in the standard question database, the user question is matched with the standard question database to determine the standard question, and a text answer template corresponding to the user question is determined through the standard question.

[0039] The question-answer template matching module checks whether the device ID exists in the pre-built standard question database. If so, the user question is matched with the question in the standard question database, and the top 1 similarity (maximum similarity) and the ID of the standard question are returned. Further, the text answer template is matched in the preset question-answer template database based on the ID of the standard question. The text answer template includes the preset text answer to the user question. The text answer template is used to provide a preliminary solution to the user question. The user question can be answered step by step according to the format of the text answer template.

[0040] like Figure 2 As shown in FIG. 1 , the standard question database includes a mapping relationship between a standard question text vector, a question ID, and a preset device ID. If the device ID belongs to a preset device ID in the standard question database, the standard question corresponding to the maximum similarity is searched in the standard question database according to the user question. The text answer template corresponding to the standard question is further found in the pre-built question-answer template database according to the standard question, for example, Figure 3 As shown, the question-answer template database includes multiple question-answer template data tables for preset device IDs. Each question-answer template data table for a preset device ID includes multiple question IDs and corresponding answer templates. The ID of the preset device is matched according to the ID of the standard question, and the answer template is further matched, so that the text answer template corresponding to the user question is finally determined through the standard question.

[0041] S300: Classify the user's question to determine the type of the user's question.

[0042] like Figure 7As shown, user questions are classified to determine the type of user questions. For example, a question classification model is constructed. (1) Collect sample user questions: accumulate sample user questions that users often ask, including actual business data, Internet data, etc. (2) Data preprocessing and question category annotation: remove special characters, punctuation marks, etc. from sample user questions and annotate the question types. For example, define a variety of basic question types, such as product information questions (for example, what is the rated power of the device?), equipment usage questions (for example, how to make an appointment to cook porridge / rice / soup?), troubleshooting questions (for example, my rice cooker display shows C2, what should I do?), equipment maintenance questions (for example, what is the warranty service of the rice cooker?), cleaning and maintenance questions (for example, how should the rice cooker be cleaned?). (3) Lexical segmentation and word embedding: decompose the sample user questions into multiple tokens, and then embed each token to convert it into a numerical form that can be learned by the preset model, and obtain numerical samples for model training. (4) Question classification model training: Obtain a preset model, for example, the preset model is a Bidirectional Encoder Representations from Transformers (BERT) model. The BERT model is a complex classification model based on transformers that can understand the contextual information of words. The BERT model is trained based on numerical samples to learn the patterns and associations between sample user questions and their question types. After the training is completed, a question classification model is obtained. The question type of the user question is automatically determined based on the question classification model.

[0043] Furthermore, user questions may be diverse and ambiguous, leading to inaccurate judgment of user types. For example, a user's question is "The rice cooker is not working well, what should I do?" In order to increase the accuracy of understanding user questions, when the question type cannot be determined, a large language model is used to determine the question type, such as using prompt words to let a more capable large language model make a judgment, or further questioning the user to determine the question type.

[0044] S400: When it is determined that the question type is a question type that needs to be answered in combination with a picture, a search is performed based on the device ID, the user question, and the text answer template, and a text analysis and a display picture are obtained to determine a solution.

[0045] Specifically: the question type output by the question classification model can be obtained. The solution is determined according to the question type. The solution includes text analysis and picture display solution (picture and text solution).

[0046] For example, if the question type is determined to be at least one of a device usage question, a troubleshooting question, or a device maintenance question, then the user type is determined to be a question type that needs to be answered with a picture. The device ID, user question, and text answer template are input into the picture-based question-answering module. The picture-based question-answering module searches based on the device ID, user question, and text answer template, obtains text analysis, and displays pictures to determine the solution.

[0047] For example, the user's question is how to use a rice cooker to steam rice? The text answer template is step one, plug in the power, step two, press the timer button, and step three, press the steam button. Search for the corresponding display image and detailed text analysis based on the device ID, user question, and text answer template. Insert the display image and detailed text analysis into the corresponding steps of the text answer template to get the solution.

[0048] The user question and answer method provided in the embodiment of the present application searches for text analysis and displays pictures through device ID, user questions and text answer templates, thereby realizing a picture-based answer. This application provides a "fool-proof" method for guiding users to use or operate a device in the field of question and answer applications for home appliances and other devices, and proposes an applicable human-computer interaction process. This application matches the text answer template according to the device ID to quickly obtain preliminary answers to user questions. Further searches are performed through the device ID and user questions to obtain solutions containing text analysis and display pictures, making the solutions more vivid, which is conducive to improving user understanding and improving the accuracy and efficiency of user questions and answers.

[0049] Based on the above embodiment, the user question and answer method further includes the following steps.

[0050] When the device ID does not exist in the standard question database, the device information corresponding to the device ID is obtained, and the device manual content is searched online based on the device information.

[0051] A first prompt instruction is generated according to the device ID and the user question, and a text analysis is obtained from the manual content of the device according to the first prompt instruction to determine a solution.

[0052] like Figure 7 As shown, the user question and answer device may also include a manual online search module and a general question and answer module. If the device ID cannot be matched in the standard question database, the device information corresponding to the device ID (for example, the device ID and the device category) is obtained. The device information is input into the manual online search module. The manual online search module searches the manual content of the device based on the device information and sends the manual content of the device to the general question and answer module.

[0053] The manual online search module relies on search engines and implements search by calling search engine application programming interfaces (APIs), such as Baidu and Google. The search results of the manual online search module include structured data, such as JSON data, and the JSON data is parsed to obtain the manual content of the corresponding device.

[0054] The general question and answer module is used for general equipment-related knowledge questions and answers. On the basis of understanding the content of the entire equipment manual, it correctly understands the user's intentions and gives comprehensive and reasonable answers to questions expressed in diverse and ambiguous language.

[0055] Considering the complexity of constructing question-answer templates and the difficulty of standard questions covering all user questions, this application has designed a more general instruction manual question-answering strategy. The general question-answering module of this application uses the reading comprehension and dialogue generation capabilities of the language large model to obtain text analysis from the device's instruction manual content and answer user questions.

[0056] The general question-answering module obtains the device ID, device question, and device manual content, and generates a first prompt instruction based on the device ID and user question through the prompting process. The first prompt instruction is input into the trained big model to stimulate the reading comprehension and conversation ability of the trained big model. The trained big model obtains text analysis from the device manual content according to the first prompt instruction to determine the solution.

[0057] For example, the first prompt instruction is as follows: 0. User question (question); 1. You are a strong assistant who can answer questions related to the manual as well as engage in small talk Q&A; 2. When the user's question is not related to the use of home appliances such as air fryers and rice cookers, ignore the content of points 3 and 4 and answer the user's question normally; 3. Otherwise, answer the question concisely and professionally based on the “device manual content”; 4. The manual content of the device (manual_content); 5. Answer directly, do not repeat the question, try to explain in paragraphs, and please use Chinese for the answer.

[0058] This application solves the problem of user questions and answers when the device ID cannot be retrieved in the standard question database by searching the device manual content online, meeting the needs of diverse device manual questions and answers. By constructing the first prompt command to automatically obtain the answer solution, a "fool-proof" method of guiding users to use or operate the device is provided, simplifying the user question and answer process.

[0059] This application provides hierarchical answers to user questions of different question types, taking into account both response speed and good accuracy.

[0060] Based on the above embodiment, the user question and answer method further includes the following steps.

[0061] When it is determined that the question type is not a question type that needs to be answered in conjunction with a picture, the device manual content of the device is obtained by matching with the device manual database based on the device ID.

[0062] A first prompt instruction is generated according to the device ID and the user question, and a text analysis is obtained from the manual content of the device according to the first prompt instruction to determine a solution.

[0063] If it is determined that the question type is not a question type that needs to be answered in combination with a picture, the device ID and the user question are sent to the general question-answering module.

[0064] The general question and answer module can also obtain the device manual content from the preset device manual database according to the device ID. Figure 4 As shown in the figure, the equipment manual database includes a one-to-one mapping relationship between the preset equipment ID and the content of the manual. The equipment manual database uses the Remote Dictionary Service (Redis) database framework. Considering the limited number of manual documents (more than 10,000) in actual business applications, all models of equipment manuals share a data table.

[0065] In order to minimize the latency of the Q&A service and effectively reduce the number of disk input / output (I / O) operations that frequently read the device manual database, the device manual database stores the device ID and the content of the corresponding manual, rather than the address of the manual document. During the initialization phase of the device manual database, the contents of all manuals are read in sequence through preloading and saved in the device manual database. In this way, when providing formal Q&A services, questions and answers can be quickly answered based on the contents of the manual, saving the time consumed in loading the manual document.

[0066] After the general question-and-answer module matches the manual content of the device, it generates a first prompt instruction according to the device ID and the user question, and obtains text analysis from the manual content of the device according to the first prompt instruction to determine a solution.

[0067] The present application obtains the device's instruction manual content through a pre-built device instruction manual database, thereby avoiding online search and improving the efficiency of obtaining the device's instruction manual content.

[0068] Based on the above embodiment, searching according to device ID, user question and text answer template, obtaining text analysis and displaying pictures to determine the solution includes the following steps.

[0069] The device ID is matched with the image database to determine the image information corresponding to the device ID, where the image information includes the image ID and the image description.

[0070] A second prompt is generated based on the image information, the user question and the text answer template.

[0071] The image information and the user question are rearranged based on the second prompt to obtain a text analysis containing the image ID.

[0072] Get the display image from the image database based on the image ID.

[0073] Get the solution based on the displayed pictures and text analysis.

[0074] like Figure 5 As shown, a picture database is pre-built, and the picture database includes a plurality of picture data tables of preset device IDs. Each picture data table of preset device ID includes a one-to-one mapping relationship between a preset picture ID, a text description and picture encoding data. The picture encoding data includes a simple markup language (markdown).

[0075] like Figure 7 As shown, the device ID is matched with the preset device ID in the image database to obtain relevant image information.

[0076] The second prompt is generated based on the image information, the user question and the text answer template. For example, the second prompt is as follows: 0. User question (question); 1. Text answer template (answer); 2. Picture information (picture_desc); 3. You are a powerful assistant. Please regenerate the answer based on the above user questions and text answer templates; 4. You need to add the image ID of the auxiliary device operation to the generated answer. You can refer to the image information in point 2. The output format is strictly required as follows; 5. The answer must be explained step by step in the format of the text answer template; 6. The image ID must be enclosed in square brackets to facilitate subsequent image restoration; 7. Answer directly, do not repeat the question, try to explain in paragraphs, and please use Chinese for your answers.

[0077] Among them, question represents the user question. answer represents the text answer template. picture_desc represents the picture ID and picture description.

[0078] The second prompt is input into the trained big model. The trained big model reorganizes the image information and the user's question to obtain a text analysis containing the image ID.

[0079] Get the display image from the image database according to the image ID, and get the solution based on the display image and text analysis.

[0080] Post-process the generated text analysis containing the picture ID, query the picture database according to the picture ID, obtain the picture encoding data corresponding to the picture ID, decode the picture encoding data, and restore the display picture of the auxiliary device operation. Arrange the display picture into the text answer template to form a picture-text solution that integrates "text guidance and auxiliary operation device pictures".

[0081] Furthermore, the trained large models include language large models, or open source language models, such as the QWen series, Chat Generation Pre-trained Transformer (ChatGPT), General Language Model (GLM) 4, etc.

[0082] The present application constructs a second prompt, obtains a text analysis containing an image ID according to the second prompt, and then obtains a solution with both pictures and texts, making the solution more vivid and conducive to "fool-proof" guidance of users to operate and use the device.

[0083] Based on the above embodiment, searching according to device ID, user question and text answer template, obtaining text analysis and displaying pictures to determine the solution includes the following steps.

[0084] Get the initial display image of the device entered by the user.

[0085] The instruction manual content of the device is searched based on the device ID, and at least one of the instruction manual content of the device and the text answer template is used as text data.

[0086] Integrate and organize user questions, initial display images, and text data, and obtain text analysis and display images to determine the solution.

[0087] The initial display picture includes the operation panel, display screen, etc. of the device. First, build an image recognition and object detection model to identify the operation buttons or fault areas of household appliances. Secondly, guide the user to take pictures to obtain the initial display picture. Through the pre-built image recognition and object detection model, the position and identification of the function buttons or interfaces of the device in the initial display picture are identified. Then, according to the device ID, the manual content and / or text answer template of the device are obtained to obtain the text data corresponding to the device ID. The user's questions, text data, and the identified initial display pictures are sent to the multimodal large model, combined with multiple modal information such as text and images for comprehensive understanding, and text analysis and display pictures are provided to provide users with corresponding solutions. The multimodal large model can be GPT-4O, GLM-4V, etc.

[0088] Furthermore, based on the real-time scene provided by the user's camera, the augmented reality (AR) engine and simultaneous localization and mapping (SLAM) technology are used to detect the device's location and spatial changes in real time through the device's environmental data collected by the camera, and virtual information (including labels, arrows, operation instructions, etc.) is superimposed on the device in the real scene. The system generates 3D models for devices or operations in advance, and these 3D models are loaded and rendered in real time according to the scene when the user requests it. For example, it shows how to set the timing of home appliances through buttons.

[0089] This application provides an efficient graphic and text-based solution by integrating and organizing user questions, initial display images and text data, obtaining text analysis and display images to determine the solution.

[0090] Based on the above embodiment, the user question is matched with the standard question database to determine the standard question, including: converting the user question into a user question text vector, obtaining the similarity between the user question text vector and each standard question text vector in the standard question database, and taking the question corresponding to the standard question text vector with the greatest similarity as the standard question.

[0091] like Figure 2As shown, the standard question database includes a plurality of standard question vector tables of preset device IDs. The standard question vector table of each preset device ID includes a one-to-one mapping relationship between a plurality of standard question text vectors and a plurality of question IDs. If the device ID belongs to a preset device ID in the standard question database, the standard question corresponding to the maximum similarity is searched in the standard question database according to the user question. According to the standard question, the text answer template corresponding to the standard question is further found in the pre-constructed question-answer template database.

[0092] A standard question database is constructed. The sources of questions include actual business data, e-commerce platform comment data, etc. The standard question database uses the open source Facebook AI Similarity Search (Faiss) vector database framework to construct a standard question vector data table with a preset device ID as the index for home appliances and other devices. The index of the standard question vector data table uses the default index ID (question ID) in the Faiss library, and the value is the standard question text vector. In the Faiss library, once the standard question vector data table is constructed, the default question ID remains fixed and does not change. Therefore, the question corresponding to the standard question text vector can be uniquely determined based on the question ID. The standard question text vector is obtained by vectorizing the question through the embedding model. For example, the vector dimension is set to 1024. The embedding model can be a pre-trained BGE Large Chinese Model (bge-large-zh) model. When matching standard questions, for a certain device ID, the user question is converted into a user question text vector. For example, the user question is converted into a user question text vector based on the embedding model to facilitate matching the standard question text vector in the standard question database. The matching algorithm adopts the classic Flat search method in Faiss, which has the advantage of high matching accuracy without the need for complex preprocessing or training. The similarity obtained by matching is a scalar. In order to reduce the delay of question and answer as much as possible, this application only matches the standard question text vector with the maximum similarity. The question corresponding to the standard question text vector with the greatest similarity is used as the standard question. According to the standard question database, the ID of the standard question is obtained. According to the ID of the standard question, the corresponding standard question text answer template is retrieved from the question-answer template database.

[0093] The question-answer template database uses the Redis database framework. The Redis service has a high-throughput I / O design and sub-millisecond operation latency, which is suitable for frequent question-answer template retrieval and real-time question and answer services.

[0094] This application obtains the standard question text vector with the greatest similarity and then obtains the standard question, thereby maximizing the reduction of question-answering delay and improving the efficiency of user question-answering.

[0095] Furthermore, the preset device IDs of the standard question database, question-answer template database, equipment manual database and image database of the present application are consistent, that is, the preset device ID in any of the above databases can be directly used for data retrieval in other databases.

[0096] The user question and answer device provided by the present application is described below. The user question and answer device described below and the user question and answer method described above can be referenced to each other.

[0097] like Figure 8 As shown, a user question-answering device includes: a device identification module 801, which is used to obtain user questions and device IDs corresponding to the user questions.

[0098] The question and answer template matching module 802 is used to match the user question with the standard question database to determine the standard question when the device ID exists in the standard question database, and determine the text answer template corresponding to the user question through the standard question.

[0099] The question type determination module 803 is used to classify the user's question to determine the question type of the user's question.

[0100] The picture and text question and answer module 804 is used to search according to the device ID, user question and text answer template when it is determined that the question type is a question and answer type that needs to be answered with a picture, obtain text analysis and display pictures to determine the solution.

[0101] The user question and answer device provided in the embodiment of the present application searches for text analysis and displays pictures through device ID, user questions and text answer templates, thereby realizing a picture-based answer. Aiming at the field of question and answer applications for home appliances and other devices, the present application provides a "fool-proof" method for guiding users to use or operate a device, and at the same time proposes an applicable human-computer interaction process. The present application matches the text answer template according to the device ID, thereby realizing the rapid acquisition of preliminary answers to user questions. Further searches are performed through the device ID and user questions to obtain solutions containing text analysis and displayed pictures, making the solutions more vivid, which is conducive to improving user understanding and improving the accuracy and efficiency of user questions and answers.

[0102] In one embodiment, the user question-and-answer device further includes a manual online search module and a general question-and-answer module. The manual online search module is used to: when the device ID does not exist in the standard question database, obtain the device information corresponding to the device ID, and search the device manual content online based on the device information. The general question-and-answer module is used to: generate a first prompt instruction based on the device ID and the user question, and obtain a text analysis from the device manual content based on the first prompt instruction to determine a solution.

[0103] In one embodiment, the picture and text question and answer module 804 is further used to: when it is determined that the question type is not a question type that needs to be answered in combination with a picture, match the device manual database based on the device ID to obtain the device manual content. The general question and answer module is used to: generate a first prompt instruction based on the device ID and the user's question, and obtain a text analysis from the device manual content according to the first prompt instruction to determine a solution.

[0104] In one embodiment, the picture-text question-answering module 804 is used to: match the device ID with the picture database to determine the picture information corresponding to the device ID, the picture information including the picture ID and the picture description; generate a second prompt based on the picture information, the user question and the text answer template; rearrange the picture information and the user question based on the second prompt to obtain a text analysis containing the picture ID; obtain a display picture from the picture database according to the picture ID; and obtain a solution based on the display picture and the text analysis.

[0105] In one embodiment, the graphic and text question and answer module 804 is used to: obtain the initial display image of the device input by the user; search the device's manual content based on the device ID, and use at least one of the device's manual content and the text answer template as text data; integrate and organize user questions, initial display images and text data, and obtain text analysis and display images to determine solutions.

[0106] In one embodiment, the question-answer template matching module 802 is used to: convert the user question into a user question text vector, obtain the similarity between the user question text vector and each standard question text vector in the standard question database, and take the question corresponding to the standard question text vector with the greatest similarity as the standard question.

[0107] Fig. 9 An example of a physical structure diagram of an electronic device is shown in FIG. Fig. 9As shown, the electronic device may include: a processor 910, a communication interface 920, a memory 930 and a communication bus 940, wherein the processor 910, the communication interface 920 and the memory 930 communicate with each other through the communication bus 940. The processor 910 may call the logic instructions in the memory 930 to execute the user question and answer method, which includes: obtaining a user question and a device ID corresponding to the user question; in the case where the device ID exists in the standard question database, matching the user question with the standard question database to determine the standard question, and determining the text answer template corresponding to the user question through the standard question; classifying the user question to determine the question type of the user question; in the case where it is determined that the question type is a question type that needs to be answered in combination with a picture, searching according to the device ID, the user question and the text answer template, obtaining text analysis and displaying pictures to determine the solution.

[0108] In addition, the logic instructions in the above-mentioned memory 930 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0109] On the other hand, the present application also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the user question and answer method provided by the above methods, which method includes: obtaining a user question and a device ID corresponding to the user question; when the device ID exists in a standard question database, matching the user question with the standard question database to determine the standard question, and determining a text answer template corresponding to the user question through the standard question; classifying the user question to determine the question type of the user question; when it is determined that the question type is a question type that needs to be answered in combination with a picture, searching based on the device ID, user question and text answer template, obtaining text analysis and displaying pictures to determine a solution.

[0110] On the other hand, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the user question and answer method provided by the above-mentioned methods, the method comprising: obtaining a user question and a device ID corresponding to the user question; in a case where the device ID exists in a standard question database, matching the user question with the standard question database to determine a standard question, and determining a text answer template corresponding to the user question through the standard question; classifying the user question to determine the type of the user question; in a case where it is determined that the question type is a type of question that needs to be answered in combination with a picture, searching based on the device ID, the user question and the text answer template, obtaining a text analysis and displaying a picture to determine a solution.

[0111] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0112] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A user question and answer method, characterized in that: include: Obtain a user question and a device ID corresponding to the user question; In the case where the device ID exists in a standard question database, matching the user question with the standard question database to determine a standard question, and determining a text answer template corresponding to the user question through the standard question; Classifying the user's problem to determine the problem type of the user's problem; When it is determined that the question type is a question type that needs to be answered in combination with a picture, a search is performed based on the device ID, the user question, and the text answer template to obtain a text analysis and display a picture to determine a solution.

2. The user question and answer method according to claim 1, characterized in that: The method further comprises: If the device ID does not exist in the standard question database, obtain device information corresponding to the device ID, and search the device manual content online according to the device information; A first prompt instruction is generated according to the device ID and the user question, and a text analysis is obtained from the instruction manual of the device according to the first prompt instruction to determine a solution.

3. The user question and answer method according to claim 1, characterized in that: The method further comprises: In the case where it is determined that the question type is not the question type that needs to be answered in combination with a picture, matching the device manual database based on the device ID to obtain the device manual content; A first prompt instruction is generated according to the device ID and the user question, and a text analysis is obtained from the instruction manual of the device according to the first prompt instruction to determine a solution.

4. The user question and answer method according to claim 1, characterized in that: The searching according to the device ID, the user question and the text answer template, obtaining text analysis and displaying pictures to determine a solution, includes: Matching the device ID with a picture database to determine picture information corresponding to the device ID, the picture information including a picture ID and a picture description; Generate a second prompt based on the picture information, the user question and the text answer template; Rearrange the picture information and the user question based on the second prompt to obtain the text analysis containing the picture ID; Acquire the display picture from the picture database according to the picture ID; The solution is obtained based on the displayed image and the text analysis.

5. The user question and answer method according to claim 1, characterized in that: The searching according to the device ID, the user question and the text answer template, obtaining text analysis and displaying pictures to determine a solution, includes: Get the initial display image of the device entered by the user; Searching for the instruction manual content of the device based on the device ID, and taking at least one of the instruction manual content of the device and the text answer template as text data; The user question, the initial display image and the text data are integrated and sorted, and the text analysis and the display image are obtained to determine the solution.

6. The user question and answer method according to claim 1, characterized in that: The step of matching the user question with the standard question database to determine a standard question includes: The user question is converted into a user question text vector, the similarity between the user question text vector and each standard question text vector in the standard question database is obtained, and the question corresponding to the standard question text vector with the greatest similarity is used as the standard question.

7. A user question and answer device, characterized in that: include: A device identification module, used to obtain a user question and a device ID corresponding to the user question; A question-answer template matching module, configured to, if the device ID exists in a standard question database, match the user question with the standard question database to determine a standard question, and determine a text answer template corresponding to the user question through the standard question; A question type determination module, used to classify the user question to determine the question type of the user question; The picture and text question and answer module is used to search according to the device ID, the user question and the text answer template when it is determined that the question type is a question and answer type that needs to be answered with the help of pictures, obtain text analysis and display pictures to determine the solution.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the user question and answer method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the user question and answer method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the user question and answer method according to any one of claims 1 to 6 is implemented.

Images