Question answering searching method and corresponding apparatus

Abstract

A question answering searching method can be applied to an end-cloud collaboration scenario, and the method comprises: a client sends a query request (query) and historical information to a cloud, and the cloud guides the generation of an answer to the query request according to the historical information, so as to obtain an answer associated with the historical information. The historical information can reflect the characteristics of users using the client, such that the generated answers are more in line with user perception, thereby improving the matching degree between answers and users.

Description

A method for searching questions and answers and corresponding device

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on August 17, 2023, with application number 202311044309.6 and application name “A method for searching questions and answers and corresponding device”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a method for searching questions and answers and a corresponding device. Background Art

[0003] Search generally refers to a retrieval technology that uses specific strategies to retrieve specified information from the Internet and feeds it back to users based on user needs and certain algorithms. Its key feature is to find the web pages / documents that are most relevant to the user's query from massive web page libraries / document libraries.

[0004] Question answering (QA) is a form of search that uses accurate and concise natural language to answer questions raised by users, satisfying people's need for fast and accurate information acquisition. Unlike search, QA directly provides answers rather than related web pages or documents.

[0005] In current search question-answering systems, the answers generated for the same query are basically the same for all users.

[0006] Summary of the Invention

[0007] This application provides a method for searching questions and answers to improve the matching degree between answers and users. This application also provides a corresponding device, a computer-readable storage medium, and a computer program product.

[0008] In a first aspect, the present application provides a method for searching questions and answers, comprising: sending a first query request and corresponding historical information to the cloud, where the historical information includes at least one of historical query requests and corresponding answers associated with the first query request; receiving a first answer from the cloud, where the content of the first answer is associated with the historical information; and presenting the first answer.

[0009] In this application, search question and answer is a search process based on giving accurate answers to questions.

[0010] In this application, the cloud can be software or services on a cloud platform, or software or services deployed on nodes in a network, such as edge nodes. The cloud can run on independent physical machines or on virtualized resources.

[0011] In this application, the client can be a terminal device or an application, for example, an application running on a terminal device for user use. The application can be a search engine, an intelligent voice assistant, an intelligent agent social network, or a variety of question-and-answer or dialogue applications.

[0012] In this application, historical information can be historical query requests sent by the user through the same terminal device or the same user account, and at least one of the answers corresponding to these historical query requests. That is to say, historical information can be just historical query requests, or just answers corresponding to historical query requests, or both historical query requests and corresponding answers.

[0013] In this application, the first answer is generated in the cloud based on historical information from the client. The historical information can reflect the characteristics of the user using the client. The first answer generated in this way is more in line with the user's perception and is a high-quality answer that conforms to the user's personality, which can improve the matching degree between the answer and the user.

[0014] In one possible implementation, before sending the first query request and corresponding historical information to the cloud, the method further includes: parsing the first query request to retrieve historical information from multiple historical query requests and corresponding answers stored on the client, where the similarity between the historical information and the first query request is greater than a first threshold.

[0015] In this possible implementation, the client may store a large amount of historical information, and much of this historical information may be irrelevant to the first query request and therefore have no reference value. Therefore, the client will filter the locally stored historical information in conjunction with the first query request. The filtering principle can also be to use cosine similarity or other distance calculation methods to determine the historical information related to the first query request. In this way, the accuracy of the historical information in guiding the first answer can be improved. Moreover, performing the filtering on the client can also enhance the privacy of user data and reduce the amount of historical information transmitted to the cloud.

[0016] In one possible implementation, the first answer is determined by the cloud based on the second query request, and the second query request is obtained by the cloud by rewriting the first query request based on historical information.

[0017] In this possible implementation, the first query request may not fully comply with the specification, making it difficult for machines to recognize it. The cloud can then rewrite the first query request based on historical information, allowing the machine to better understand the search content entered by the user. This can improve the efficiency and quality of search and answering.

[0018] In a possible implementation, the first query request also includes a first style selected by the user, and the first style is used to indicate the presentation form of the first answer; the above step of presenting the first answer includes presenting the content of the first answer in the first style.

[0019] In a possible implementation, the first style is selected by the user from multiple styles through a style button on the search question and answer interface, and multiple styles are displayed after the style button is triggered.

[0020] In the above possible implementation methods, the client may include a setting for a style button, under which a variety of styles may be associated, for example, rigorous, heartwarming, and two-dimensional styles. The user can select the style in which the answer is presented from the style options displayed for the various styles. In this way, the client will carry the first style when sending the first query request, and the cloud will combine the first style with the content of the first answer to obtain the first answer. After receiving the first answer, the client will present the content of the first answer in the first style. Optionally, after the user selects the first style, the client may not carry the first style in the first query request, but after receiving the content of the first answer, adjust the content of the first answer according to the first style and then present it.

[0021] In this application, presenting the content of the answer according to the style selected by the user can improve the matching degree between the answer presentation format and the user's preferences, thereby increasing user stickiness.

[0022] In one possible implementation, the first answer includes a second style, and the second style is a presentation form of the first answer determined by the cloud based on historical information; the above step: presenting the first answer includes: presenting the content of the first answer in the second style.

[0023] In this possible implementation, the cloud can also summarize user preferences based on historical information, and then determine the presentation style of the answer based on the user's preferences, such as: a rigorous style or a two-dimensional style, etc. When the user does not specify the presentation style of the answer, the first answer will be generated based on the second style determined by the historical information. This can improve the matching degree between the answer presentation form and the user's preferences, thereby increasing user stickiness.

[0024] In one possible implementation, if the first query request conflicts with at least one of the historical query requests, a clarification response is sent to the cloud, where the clarification response is used to instruct the cloud to ignore the conflicting historical query requests, or the clarification response is used to instruct the cloud not to ignore the conflicting historical query requests.

[0025] In this possible implementation, the conflict judgment can be notified to the client in the form of a query clarification request after the cloud makes a judgment, or it can be judged by the client itself, and this application does not limit this. The query clarification request refers to the conflict between the content requested by the first query request this time and one or more requested contents in the historical query requests. For example, a historical query request requests content about rheumatoid arthritis, and the first query request this time requests content about swimming. Because swimming is likely to cause rheumatoid arthritis, the two query requests conflict with each other. The cloud can inform the user of the conflict by sending a query clarification request to the client, and ask whether to ignore the historical query request. This solution can ask the user in the form of query clarification when there is a conflict between the current query request and the historical query request, rather than making overly intelligent decisions, which can improve the user experience.

[0026] In one possible implementation, if the clarification response is used to instruct the cloud to ignore the conflicting historical query request, the first answer does not include information associated with the conflicting historical query request.

[0027] In this possible implementation, if the user confirms to ignore, the first answer generated by the cloud will only contain content corresponding to the first query request and will not include information associated with conflicting historical query requests. For example, if the first query request is about swimming, the first answer will only contain content related to swimming and will not include content related to rheumatoid arthritis.

[0028] In a possible implementation, if the clarification response is used to instruct the cloud to not ignore conflicting historical query requests, the first answer includes conflict prompt information, where the conflict prompt information is based on information determined by the conflicting historical query requests and the first query request.

[0029] In this possible implementation, if the user confirms not to ignore it, the first answer generated in the cloud will include information determined based on the conflicting historical query request and the first query request. For example, if the first query request is about swimming, the first answer generated will include relevant content about paying attention to rheumatoid arthritis when swimming.

[0030] In one possible implementation, the method further includes: responding to a user's selection of a memory mode, determining corresponding historical information for the first query request, and presenting the memory mode in the form of a button on the search question-and-answer interface.

[0031] In this possible implementation, the client's search and answer interface can be configured with "memory mode" and "forgotten mode." If the user selects forgotten mode, the client will not send historical information, and the cloud will only need to determine the corresponding answer based on the first query request. If the user selects for memory mode, the client will determine the corresponding historical information for the first query request. This can improve the accuracy of the historical information in guiding the first answer.

[0032] The second aspect of the present application provides a method for searching questions and answers, including: receiving a first query request and corresponding historical information from a client, the historical information including at least one of the historical query requests and corresponding answers associated with the first query request; generating a first answer based on the first query request and the historical information, the content of the first answer being associated with the historical information; and sending the first answer to the client.

[0033] In this application, the cloud generates a first answer to the first query request through guidance from historical information from the client. The historical information can reflect the characteristics of the user using the client, so that the first answer generated in this way is more in line with the user's perception and can improve the matching degree between the answer and the user.

[0034] In one possible implementation, generating the first answer based on the first query request and historical information includes: rewriting the first query request based on the historical information to obtain a second query request; and generating the first answer based on the second query request and historical information.

[0035] In this possible implementation, the first query request may not fully comply with the specification, making it difficult for machines to recognize it. The cloud can then rewrite the first query request based on historical information, allowing the machine to better understand the content the user is searching for in the first query request. This can improve the efficiency and quality of search and answering questions.

[0036] In one possible implementation, the first query request also includes a first style selected by the user, and the first style is used to indicate the presentation form of the first answer; the above step: generating the first answer based on the second query request and historical information includes: determining the content of the first answer based on the second query request and historical information, and integrating the first style with the content of the first answer to obtain the first answer.

[0037] In this possible implementation, the cloud combines the first style with the content of the first answer to generate the first answer. After receiving the first answer, the client then presents the first answer content in the first style. This way, presenting the answer content in the style selected by the user can better match the answer presentation format with the user's preferences, thereby increasing user stickiness.

[0038] In one possible implementation, the above steps of generating a first answer based on the second query request and historical information include: determining the user's style preference based on the historical information, and determining a second style based on the style preference, the second style being used to indicate the presentation form of the first answer; determining the content of the first answer based on the second query request and historical information, and integrating the second style with the content of the first answer to obtain the first answer.

[0039] In this possible implementation, the cloud can also summarize user preferences based on historical information, and then determine the presentation style of the answer based on the user's preferences, such as: a rigorous style or a two-dimensional style, etc. When the user does not specify the presentation style of the answer, the first answer will be generated based on the second style determined by the historical information. This can improve the matching degree between the answer presentation form and the user's preferences, thereby increasing user stickiness.

[0040] In one possible implementation, the method further includes: determining whether there is a conflict between the first query request and the historical query request; if there is a conflict, sending a query clarification request to the client, the query clarification request is used to indicate that there is a conflict between the first query request and at least one of the historical query requests, and is used to inquire whether to ignore the conflicting historical query request; receiving a clarification response from the client, the clarification response is used to indicate to ignore the conflicting historical query request, or not to ignore the conflicting historical query request.

[0041] In this possible implementation, when a current query request conflicts with a historical query request, the user may be asked for query clarification instead of making an overly intelligent decision, thereby improving user experience.

[0042] In one possible implementation, if the clarification response is used to instruct the cloud to ignore conflicting historical query requests, the conflicting historical query requests and the corresponding answers are deleted from the historical information before generating the first answer based on the first query request and the historical information; accordingly, the first answer does not contain information associated with the conflicting historical query requests.

[0043] In a possible implementation, if the clarification response is used to instruct the cloud not to ignore conflicting historical query requests, then correspondingly, the first answer includes conflict prompt information, which is based on information determined by the conflicting historical query requests and the first query request.

[0044] In a third aspect, the present application provides a client for executing the method of the first aspect or any possible implementation of the first aspect. Specifically, the client includes modules or units for executing the method of the first aspect or any possible implementation of the first aspect, such as a receiving unit, a display unit, and a sending unit.

[0045] In a fourth aspect of the present application, a cloud device is provided for executing the method of the second aspect or any possible implementation of the second aspect. Specifically, the cloud device includes modules or units for executing the method of the second aspect or any possible implementation of the second aspect, such as a processing unit, a sending unit, and a receiving unit.

[0046] In a fifth aspect, the present application provides a client comprising a transceiver, a processor and a memory, wherein the transceiver and the processor are coupled to the memory, and the memory is used to store programs or instructions. When the programs or instructions are executed by the processor, the cloud device executes the method in the aforementioned first aspect or any possible implementation of the first aspect.

[0047] In a sixth aspect of the present application, a cloud device is provided. The cloud device may include at least one processor, a memory, and a communication interface. The processor is coupled to the memory and the communication interface. The memory is configured to store instructions, the processor is configured to execute the instructions, and the communication interface is configured to communicate with other network elements under the control of the processor. When executed by the processor, the instructions cause the processor to perform the method of the second aspect or any possible implementation of the second aspect.

[0048] In the seventh aspect of the present application, a chip system is provided, which includes one or more interface circuits and one or more processors; the interface circuits and the processors are interconnected by lines; the interface circuits are used to receive signals from the client's memory and send signals to the processor, and the signals include computer instructions stored in the memory; when the processor executes the computer instructions, the client executes the method in the aforementioned first aspect or any possible implementation of the first aspect.

[0049] In an eighth aspect of the present application, a chip system is provided, which includes one or more interface circuits and one or more processors; the interface circuits and the processors are interconnected by lines; the interface circuits are used to receive signals from the memory of a cloud device and send signals to the processor, the signals including computer instructions stored in the memory; when the processor executes the computer instructions, the cloud device executes the method in the aforementioned second aspect or any possible implementation of the second aspect.

[0050] The ninth aspect of the present application provides a computer-readable storage medium having a computer program or instruction stored thereon. When the computer program or instruction runs on a computer device, the computer device executes the method in the aforementioned first aspect or any possible implementation of the first aspect.

[0051] The tenth aspect of the present application provides a computer-readable storage medium having a computer program or instruction stored thereon. When the computer program or instruction runs on a computer device, the computer device executes the method in the aforementioned second aspect or any possible implementation of the second aspect.

[0052] In an eleventh aspect, the present application provides a computer device program product, which includes a computer device program code. When the computer device program code is executed on a computer device, the computer device executes the method in the aforementioned first aspect or any possible implementation of the first aspect.

[0053] The twelfth aspect of the present application provides a computer device program product, which includes a computer device program code. When the computer device program code is executed on a computer device, the computer device executes the method in the aforementioned second aspect or any possible implementation of the second aspect.

[0054] In the thirteenth aspect of the present application, a search question and answer system is provided, which includes a cloud device and a client, wherein the client is used to execute the method in the aforementioned first aspect or any possible implementation of the first aspect, and the cloud device is used to execute the method in the aforementioned second aspect or any possible implementation of the second aspect.

[0055] Among them, the technical effects brought about by the third to thirteenth aspects or any possible implementation methods thereof can refer to the technical effects brought about by the first aspect or different possible implementation methods of the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] FIG1A is a schematic diagram of an architecture of a search question-answering system provided in an embodiment of the present application;

[0057] FIG1B is a schematic diagram of an architecture of a search question-answering system provided in an embodiment of the present application;

[0058] FIG2A is a schematic structural diagram of a terminal device provided in an embodiment of the present application;

[0059] FIG2B is a schematic diagram of a structure of a cloud device provided in an embodiment of the present application;

[0060] FIG3 is a schematic diagram of an embodiment of a method for searching questions and answers provided in an embodiment of the present application;

[0061] FIG4 is a schematic diagram of an example of a question-and-answer search page of a client provided in an embodiment of the present application;

[0062] FIG5 is a schematic diagram of an architecture of a search question and answer provided in an embodiment of the present application;

[0063] FIG6 is another schematic diagram of the architecture of the search question and answer provided in an embodiment of the present application;

[0064] FIG7 is another example schematic diagram of a question-and-answer search page of a client provided in an embodiment of the present application;

[0065] FIG8 is another schematic diagram of the architecture of the search question and answer provided in an embodiment of the present application;

[0066] FIG9 is another schematic diagram of the architecture of the search question and answer provided in an embodiment of the present application;

[0067] FIG10 is another example schematic diagram of a question-and-answer search page of a client provided in an embodiment of the present application;

[0068] FIG11 is a schematic structural diagram of a client provided in an embodiment of the present application;

[0069] FIG12 is a schematic structural diagram of a cloud device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0070] The following describes the embodiments of the present application in conjunction with the accompanying drawings. Obviously, the embodiments described are only part of the embodiments of the present application, rather than all the embodiments. Those skilled in the art will appreciate that with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0071] The terms "first," "second," and the like in the specification and claims of this application and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions, e.g., a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0072] The present application provides a method for searching questions and answers, which is used to improve the matching degree between answers and users. The present application also provides a corresponding device, a computer-readable storage medium, and a computer program product. These are described in detail below.

[0073] To facilitate understanding, the following briefly introduces the technical terms involved in the embodiments of this application:

[0074] 1. Query request (Query): The query term entered by the user in the search box.

[0075] 2. Information retrieval: Select the candidate set of documents that are most relevant to the query request from a large-scale web document library (the document library can be a library of hundreds of billions of documents).

[0076] 3. Search question-answering system: Generally refers to a system that searches for relevant documents and then answers user questions based on the understanding of the document information.

[0077] 4. Conversational search and question-answering system: This system adds multi-round conversational capabilities to the search and question-answering system, can understand the context, and thus implement multi-round continuous question-answering functions.

[0078] 5. Answer generation: Generate answers to user questions through certain methods or models.

[0079] 6. Retrieval-based answer generation: By searching the already constructed question-answer database, use the query request to match the most relevant question or answer in the question-answer database to obtain the answer.

[0080] 7. Reading comprehension-based answer generation: This method retrieves relevant web pages or documents, understands the document content, and integrates it into answers. This method can generally be divided into extractive and generative methods. The extractive method is characterized by the fact that the answer comes from web page fragments, while the generative method does not necessarily come entirely from web page document fragments and requires further processing and polishing.

[0081] 8. Large Model Generation: After training, the large model internalizes knowledge into model parameters, allowing it to directly generate answers to user queries. When using the large model to generate answers, mainstream methods can be used to guide the large model by providing prompts and generating relevant content.

[0082] 9. Historical information: query requests sent by a client in the past and / or the answers corresponding to the query requests.

[0083] 10. Historical information compression perception: Highly compress the user's historical information, save key information and enable the model to have the ability to perceive this information.

[0084] 11. User Preference: The rational and biased choices made by users when considering goods and services are the combined result of user cognition, psychological feelings and rational economic trade-offs.

[0085] 12. Query clarification: Query clarification asks relevant questions based on the query given by the user, interacts with the user through clarification or counter-questions to better understand the user's intention.

[0086] FIG1A is a schematic diagram of the architecture of a search question-answering system provided in an embodiment of the present application.

[0087] As shown in Figure 1A, the search question-and-answer system provided in an embodiment of the present application includes a cloud and multiple clients, and the cloud can communicate with multiple clients through a network. Among them, the cloud can be software or services of a cloud platform, or software or services deployed on nodes in the network, such as edge nodes. The cloud can run on an independent physical machine or on virtualized resources. The client can be a terminal device or an application, for example, the application runs on a terminal device for user use. The application can be a search engine, an intelligent voice assistant, an intelligent social agent, and other question-and-answer services, or a dialogue application.

[0088] The client can send a query request and historical information related to the query request to the cloud. The cloud-side device in the cloud can generate a corresponding answer based on the query request and historical information, and then send the answer to the client. The client receives and presents the answer.

[0089] Among them, historical information can be historical query requests sent by the user through the same terminal device or the same user account, and at least one of the answers corresponding to these historical query requests. That is to say, historical information can be only historical query requests, or only answers corresponding to historical query requests, or both historical query requests and corresponding answers.

[0090] The cloud-side device in the cloud in Figure 1A can be a working node or a scheduling node in the cloud system. After the scheduling node receives a query request from the client, the scheduling node can execute the corresponding search question and answer process. The scheduling node can also assign the query request to one or more working nodes in the cloud system, and one or more working nodes can execute the corresponding search question and answer process.

[0091] The function of the scheduling node can be implemented by software or hardware.

[0092] As an example of a software functional unit, a scheduling node may include code running on a computing instance. The computing instance may include at least one of a physical host (computing device), a virtual machine, and a container. Furthermore, the computing instance may be one or more. For example, a scheduling node may include code running on multiple hosts / virtual machines / containers. It should be noted that the multiple hosts / virtual machines / containers used to run the code may be distributed in the same region or in different regions. Furthermore, the multiple hosts / virtual machines / containers used to run the code may be distributed in the same availability zone (AZ) or in different AZs, each AZ including one data center or multiple geographically close data centers. Typically, a region may include multiple AZs.

[0093] Similarly, multiple hosts / virtual machines / containers running the code can be distributed within the same virtual private cloud (VPC) or across multiple VPCs. Typically, a VPC is set up within a region. Inter-region communication between two VPCs within the same region, or between VPCs in different regions, requires a communication gateway within each VPC to interconnect the VPCs.

[0094] As an example of a hardware functional unit, a scheduling node may include at least one computing device, such as a server. Alternatively, the scheduling node may be implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD). The PLD may be implemented using a complex programmable logical device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0095] The multiple computing devices included in a scheduling node can be distributed in the same zone or in different zones. The multiple computing devices included in a scheduling node can be distributed in the same AZ or in different AZs. Similarly, the multiple computing devices included in a scheduling node can be distributed in the same VPC or in multiple VPCs. The multiple computing devices can be any combination of servers, ASICs, PLDs, CPLDs, FPGAs, GALs, and other computing devices.

[0096] A worker node can be a physical machine, a virtual machine (VM), or a container. It can include one or more central processing units (CPUs) and graphics processing units (GPUs). A worker node can also be a CPU or a GPU.

[0097] The client can be a terminal device. A terminal device, also known as user equipment (UE), mobile station (MS), mobile terminal (MT), etc., is a device that includes wireless communication capabilities (providing voice / data connectivity to users), such as a handheld device with wireless connection capabilities. Currently, some examples of terminal devices include: mobile phones, tablets, laptops, PDAs, notebook computers, wireless routers, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving cars, wireless terminals in the Internet of Vehicles, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, or wireless terminals in smart homes. For example, a wireless terminal in the Internet of Vehicles can be an onboard device, a complete vehicle device, an onboard module, a vehicle, etc. Wireless terminals in industrial control can be robots, for example. For example, wireless terminals in unmanned driving can be drones. These terminal devices can run Android, iOS, Windows, or other operating systems. These terminal devices can run applications that require rendering of application scenes to generate two-dimensional images, such as gaming applications, lock screen applications, or map applications.

[0098] The cloud in Figure 1A above may be located in a cloud system. The architecture of the cloud system can be understood with reference to Figure 1B . As shown in Figure 1B , the cloud system includes a cloud platform and basic resources. The cloud platform includes a cloud platform manager. The scheduling node described above may be the cloud platform manager in Figure 1B . Basic resources may include multiple servers, each of which may be a worker node, or each server may include multiple worker nodes.

[0099] The working node in FIG1B may be a computing device card or a virtual machine (VM), wherein the computing device card may be at least one of a central processing unit (CPU), a graphics processing unit (GPU), and a neural network processing unit (NPU).

[0100] The cloud platform manager maintains or regularly collects information about each worker node in the underlying resources, such as resource usage (resource utilization rate or resource idle rate) on each worker node. This information can be used as auxiliary decision-making information when allocating query requests.

[0101] The cloud platform manager can receive query requests from the client, and then the cloud platform manager can execute the corresponding search question and answer process. The cloud platform manager can also assign the query request to one or more working nodes in the cloud system, and one or more working nodes can execute the corresponding search question and answer process.

[0102] After the worker nodes complete the search question and answer, the cloud platform manager can return the answer to the client.

[0103] The above-mentioned client takes a terminal device as an example. The structure of the terminal device provided in the embodiment of the present application can be understood by referring to FIG. 2A below, and the structure of the cloud device in the cloud can be understood by referring to FIG. 2B below.

[0104] Please refer to Figure 2A, which is a schematic diagram of the structure of a terminal device provided in an embodiment of the present application. As shown in Figure 2A, the terminal device may include a processor 101, a transceiver 102, a memory 103, a display 104, and a bus 105. The processor 101, the transceiver 102, the memory 103, and the display 104 are interconnected via the bus 105. In an embodiment of the present application, the processor 101 is used to control and manage the actions of the terminal device 10, for example, the processor 101 is used to respond to the process of the user inputting a query request. The transceiver 102 is used to support the communication of the terminal device 10, for example: the transceiver 102 can execute the steps of sending a query request and receiving an answer. The memory 103 is used to store the program code and data of the terminal device 10, and the display 104 is used to display the answer.

[0105] The processor 101 may be a central processing unit (CPU), a general-purpose processor (GPPC), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device (PLD), a transistor logic device (TLD), a hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. A processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The bus 105 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, for example. Buses may be classified as address buses, data buses, control buses, and the like. For ease of illustration, FIG. 2A shows only one thick line, but this does not imply that there is only one bus or only one type of bus.

[0106] FIG2A above introduces the structure of the terminal device. The structure of the cloud device will be introduced below in conjunction with FIG2B .

[0107] Figure 2B is a schematic diagram of a possible logical structure of a cloud device provided in an embodiment of the present application. As shown in Figure 2B, the cloud device 20 provided in an embodiment of the present application includes: a processor 201, a communication interface 202, a memory 203 and a bus 204. The processor 201, the communication interface 202 and the memory 203 are interconnected via the bus 204. In an embodiment of the present application, the processor 201 is used to control and manage the actions of the cloud device 20. For example, the processor 201 is used to generate answers based on query requests and historical information. The communication interface 202 is used to support the cloud device 20 to communicate. For example, the communication interface 202 can execute the steps of receiving query requests and historical information, and the step of sending answers. The memory 203 is used to store the program code and data of the cloud device 20.

[0108] The processor 201 may be a central processing unit (CPU), a general-purpose processor (GPOR), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic device (PLD), a transistor logic device (TLD), a hardware component, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. A processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like. The bus 204 may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, for example. Buses may be classified as address buses, data buses, control buses, and the like. For ease of illustration, FIG. 2B shows only one thick line, but this does not imply that there is only one bus or only one type of bus.

[0109] The following describes the method for searching questions and answers provided in an embodiment of the present application. The content related to cloud execution in this method can be executed by the cloud or by components of the cloud (such as a processor, chip, or chip system). The content related to client execution can be executed by the client or by components of the client (such as a processor, chip, or chip system).

[0110] FIG3 is a schematic diagram of an embodiment of a method for searching questions and answers provided in an embodiment of the present application.

[0111] As shown in FIG3 , an embodiment of the method for searching questions and answers provided in an embodiment of the present application includes:

[0112] 301. The client sends a first query request and corresponding historical information to the cloud. Correspondingly, the cloud receives the first query request and corresponding historical information from the client.

[0113] The history information of the first query request includes at least one of historical query requests associated with the first query request and corresponding answers.

[0114] 302. The cloud generates a first answer based on the first query request and historical information.

[0115] The content of the first answer is associated with the historical information of the first query request.

[0116] The process of generating the first answer in step 302 can be through search-based answer generation, reading comprehension-based answer generation, or large-scale model generation. For an introduction to these answer generation methods, please refer to the previous technical terminology section for an understanding. Regardless of the answer generation method used, the historical information corresponding to the first query request is used to guide the generation of the first answer.

[0117] 303. The cloud sends the first answer to the client. Correspondingly, the client receives the first answer from the cloud.

[0118] 304. The client presents the first answer.

[0119] In this application, the first answer is generated in the cloud based on historical information from the client. The historical information can reflect the characteristics of the user using the client. The first answer generated in this way is more in line with the user's perception and is a high-quality answer that conforms to the user's personality, which can improve the matching degree between the answer and the user.

[0120] The above step 304 can present the content of the first answer in a certain style. The style of presenting the content of the first answer can be selected by the user on the client or determined by the cloud based on historical information. They are introduced below respectively.

[0121] 1. The user selects a style;

[0122] The first query request also includes a first style selected by the user, where the first style is used to indicate the presentation form of the first answer. In this case, step 304 may be to present the content of the first answer in the first style.

[0123] The client may include a style button. As shown in FIG4 , a “Generate Style” button may be configured on the search question and answer page. By clicking the “Generate Style” button, multiple style options may be displayed on the same screen, such as rigorous, heartwarming, and two-dimensional styles. It is also possible to display only multiple styles without displaying the “Generate Style” button. The user can select the style in which the answer is presented on this style option. In this way, the client will carry the first style when sending the first query request, and the cloud will combine the first style with the content of the first answer to obtain the first answer. After receiving the first answer, the client will present the content of the first answer in the first style. Optionally, after the user selects the first style, the client may not carry the first style in the first query request, but after receiving the content of the first answer, adjust the content of the first answer according to the first style and then present it.

[0124] In this application, presenting the content of the answer according to the style selected by the user can improve the matching degree between the answer presentation format and the user's preferences, thereby increasing user stickiness.

[0125] 2. Determine the style in the cloud;

[0126] The first answer includes a second style, which is the presentation form of the first answer determined by the cloud based on historical information. In this case, step 304 may include presenting the content of the first answer in the second style.

[0127] The cloud can also summarize user preferences based on historical information, and then determine the presentation style of the answer based on the user's preferences, such as: a rigorous style or a two-dimensional style, etc. When the user does not specify the presentation style of the answer, the first answer will be generated based on the second style determined by the historical information. This can improve the matching degree between the answer presentation form and the user's preferences, thereby increasing user stickiness.

[0128] The processing flow of the search question and answer in the embodiment of the present application can also be understood by referring to the system architecture diagram of Figure 5. As shown in Figure 5, the system architecture of the search question and answer includes a client and a cloud.

[0129] Among them, the client can respond to the query request (query) input by the user and determine the corresponding historical information for the query request. Considering that the client may store a lot of historical information, and a lot of historical information may be irrelevant to the query request and have no reference value. Therefore, the client will filter the historical information stored locally in combination with the query request. The process can be: the client parses the first query request to retrieve historical information from multiple historical query requests and corresponding answers stored by the client, and the similarity between the historical information and the first query request is greater than a first threshold. The principle of screening can be to use cosine similarity or other distance calculation methods to determine the historical information related to the first query request, such as: historical information with a similarity greater than the first threshold. The client then sends both the query request and the filtered historical information to the cloud.

[0130] The cloud can include a query rewriting unit, a historical information processing unit, and a search question-and-answer system. The cloud can also include a personalization processing unit. The query rewriting unit can rewrite queries, potentially converting non-standard queries into machine-understandable queries. For example, if a user last entered a query on the client asking "What's the weather in city A?" and now enters a query asking "What's the weather in city B?", the query rewriting unit can rewrite the query "What's the weather in city B?" into "What's the weather in city B?" based on the information extracted from the historical information by the historical information processing unit.

[0131] The query rewriting unit transmits the rewritten query to the search question-answering system. Furthermore, the historical information processing unit can extract information strongly relevant to the query from the historical information and input it into the search question-answering system, or transmit this information to the personalization processing unit. For example, if the query is "What's the weather like in City B?", information strongly relevant to this query could include travel tips for City B, tourist attractions in City B, and hotels in City B.

[0132] The personalization processing unit can abstract the user portrait from the historical information or the information transmitted by the historical information processing unit, determine the user preferences, and then determine the style of answer presentation based on the user preferences, and transmit the style to the search question and answer system.

[0133] When generating an answer for a rewritten query, the search question-answering system combines the information transmitted by the historical information processing unit with the answer presentation style transmitted by the personalization processing unit. The cloud then sends the generated answer to the client.

[0134] The client will then present the received answer.

[0135] This process describes a situation where the answer presentation style is determined by the cloud. If the user sets the answer presentation style on the client, the personalized processing unit on the cloud may not transmit the determined style to the search question-answering system.

[0136] As shown in FIG6 , the client may include a history retriever, and the history information processing unit in FIG5 may include a memory conflict identification unit and a query clarification unit.

[0137] The history retriever can retrieve information that matches the current query (first query request) from the historical information and context, and then transmit the historical information that matches the current query to the cloud.

[0138] The memory conflict identification unit can determine whether the current query (the first query request, or the rewritten second query request, the first query request is used as an example below) conflicts with the user history. If there is a conflict, the query clarification unit is notified.

[0139] The query clarification unit may send a query clarification request to the client, the query clarification request being used to indicate that a first query request conflicts with at least one of the historical query requests and requesting whether to ignore the conflicting historical query request. After receiving the query clarification request, the client may send a clarification response to the cloud, the clarification response being used to indicate whether to ignore the conflicting historical query request or not ignore the conflicting historical query request. The query clarification unit may ignore the conflicting historical query request or not ignore the conflicting historical query request based on the clarification response.

[0140] The query clarification unit transmits the clarification response to the search question-answering system, which then generates the corresponding answer.

[0141] A query clarification request means that the content requested by the current first query request conflicts with one or more requested contents in the historical query requests. For example, a historical query request requests content about rheumatoid arthritis, while the current first query request requests content about swimming. Since swimming is likely to cause rheumatoid arthritis, the two query requests conflict with each other.

[0142] The cloud can notify the user of the conflict by sending a query clarification request to the client, and inquire whether to ignore the historical query request. If the first query request conflicts with at least one of the historical query requests, the client sends a clarification response to the cloud, which instructs the cloud to ignore the conflicting historical query request, or alternatively, instructs the cloud to not ignore the conflicting historical query request.

[0143] If the clarification response instructs the cloud to ignore the conflicting historical query, the first answer will not include information associated with the conflicting historical query. In other words, if the user confirms the "ignore" option, the first answer generated by the cloud will only contain content corresponding to the first query and will not include information associated with the conflicting historical query. For example, if the first query is about swimming, the first answer will only contain content related to swimming and will not include content related to rheumatoid arthritis.

[0144] If the clarification response instructs the cloud to not ignore the conflicting historical query, the first answer will include conflict information based on the conflicting historical query and information determined in the first query. In other words, if the user confirms that they do not want to ignore the conflicting historical query, the first answer generated by the cloud will include information determined based on the conflicting historical query and the first query. For example, if the first query is about swimming, the first answer generated will include information about the risk of rheumatoid arthritis caused by swimming.

[0145] The solution provided by the embodiment of the present application can ask the user through query clarification when there is a conflict between the current query request and the historical query request, rather than making overly intelligent decisions, which can improve the user experience.

[0146] In the content introduced in Figure 6 above, the history retriever is located on the client side, and the history retriever can retrieve the historical information associated with the first query request from the large amount of historical information stored on the client side and send it to the cloud. In another possible embodiment, the history retriever can also be located in the cloud. If the user stores all the relevant historical information that has been queried in the cloud, then the history retriever located in the cloud can query the relevant historical information based on the first query request, and then perform search and answer. In particular, when a user has multiple terminal devices, the user can log in to the applications on different terminal devices through the same user account, and the historical information queried for these terminal devices can be stored in the cloud through the user account association. In this way, when a user uses a certain terminal device, the history retriever in the cloud can filter from the historical information of the user's multiple different terminal devices to obtain the historical information related to the current query request, and then perform the search and answer process.

[0147] The above-described contents are all processes for generating answers corresponding to query requests based on historical information. Whether to use historical information in the process of generating answers can be set on the client. As shown in Figure 7, a "Memory Settings" function item can be configured on the search question and answer page of the client. When the user clicks on the "Memory Settings" function item, two modes, "Amnesia" and "Memory", will be displayed. If the user does not want historical information to be involved in answer generation, the user can select the "Amnesia" mode. If the user wants historical information to be involved in answer generation, the user can select the "Memory" mode. If the user selects the Memory mode, the client will generate the corresponding historical information for the query request.

[0148] The cloud can perform gate management based on the client's "memory settings" to determine whether to use historical information to guide answer generation. The gate management mechanism can be configured on the historical information processing unit.

[0149] The historical information processing unit can have the following functions:

[0150] 1. Historical Information Compressed Sensing: User preferences are summarized based on historical information, and representative queries are filtered out as user profiles and stored in the cloud. These representative queries (user profiles) can serve as the retrieval data source for the historical searcher, significantly compressing information capacity and preserving key user preference information.

[0151] Through multi-task joint learning, as shown in Figure 8, based on historical query requests (query) and retrieved documents (doc), the encoder captures query requests and text features to generate user labels. Under the memory setting, a gating mechanism (gate) determines whether to use historical information to guide answer generation. If the "memory setting" determines not to use historical information (amnesia mode), the search question-answering system (generator) generates answers directly. If the "memory setting" determines to use historical information (memory mode), the historical information is transmitted to the search question-answering system through the "gate", and the search question-answering system generates answers based on the historical information.

[0152] 2. Clarification of historical memory behavior conflicts: The memory conflict identification unit compares the current query with the historical profile. If the current query does not match the historical memory behavior (such as if someone else has used it), the query can be clarified (the process of querying a city area can be understood by referring to the introduction in Figure 6) to avoid excessive intelligence.

[0153] 3. Explainability of historical memory: In the references, mark the references to historical memory and explain some of the reasons for the generation of the answer.

[0154] In addition, the personalized processing unit in the embodiment of the present application can summarize the user style and develop the personality of the style. In this way, the search question and answer system can generate answers based on the user's preferred style. The process can be understood by referring to Figure 9. As shown in Figure 9, the personalized processing unit includes a style / personality development unit and a style selection / recommendation unit. If the cloud determines through memory settings that historical information needs to be used to generate answers, the style / personality development unit can be based on a multi-style text training base model, absorb the data increment training update parameters of the user's historical information, and continuously reshape the "self" style; the style / personality development unit can be a neural network model, which can summarize one or more styles based on historical information, such as one or more possible styles such as the second dimension, heartwarming, rigorous or self as shown in Figure 9. The style selection / recommendation unit can then transmit the user's preferred style summarized by the style / personality development unit to the search question and answer system, so that the search question and answer system can generate answers based on the user's preferred style.

[0155] The above describes the process of generating answers based on historical information in the cloud. The following is an example of the client-side display page, combined with the client's search question and answer page.

[0156] It should be noted that in order to facilitate understanding of the impact of historical information on this query, the answers to historical queries and the answers to this query are displayed on one page. In actual scenarios, these contents may not be displayed on one page. Of course, they can also be displayed on one page.

[0157] As shown in FIG10 , on the question-and-answer search page of the client, the user can search for “I have rheumatoid arthritis and I need to pay attention to it” in the history. If the user does not select the generated style, the answer to part 10A in FIG10 will be obtained, “Rheumatoid arthritis is a chronic disease. You need to pay attention to your diet. Do not eat raw or cold food. Eat mainly vegetarian food. Try to avoid sleeping in humid places. Do not take a shower immediately after sweating. Keep warm, strengthen physical exercise, and eat well. You can slowly adjust and cure it. You can go to the hospital for treatment and consult a doctor to learn more about rheumatoid arthritis.”

[0158] If the user selects the "rigorous" style, the answer to part 10B in Figure 10 can be obtained: "Rheumatoid arthritis patients need to pay attention to observing the condition, timely review, reasonable diet, correct exercise, and reasonable use of medication, etc."

[0159] 1. Observe the condition: Even after rheumatoid arthritis is treated and the symptoms have been alleviated, the condition may still relapse... (not fully shown in Figure 10).

[0160] In addition, Figure 10 does not fully display the details of timely review, reasonable diet, correct exercise, and reasonable use of medication. More content will be displayed in the actual search and question-and-answer scenario.

[0161] The query input by the user in this search question is "What should I wear when traveling to Xiamen in summer?" If the user selects the "Amnesia" mode in the "Memory Settings" function item, the answer to part 10C in Figure 10 will be obtained, "When traveling to Xiamen in summer, it is recommended to wear comfortable and easy-to-wash clothes, loose clothes that are not easy to wrinkle, and avoid wearing heavy clothes. Wear a thick coat when going out to prevent the weather from suddenly cooling down. If you are traveling there in August, be sure to pay more attention to local weather information, prepare a mask, wear comfortable sports shoes or travel shoes, and do not wear high heels to avoid affecting your travel."

[0162] If the user selects the "Memory" mode and the "2D" style, or if the user does not select the "2D" style, but the cloud derives the 2D style based on the previous historical information about rheumatoid arthritis, the answer to part 10D in Figure 10 will be "Casual T-shirt + shorts Xiamen is a very comfortable city, and casual summer clothes are essential. The weather is mild, neither cold nor hot, and casual summer clothes paired with slippers are very convenient for going to the sea. If you have rheumatoid arthritis, you can bring cold-proof knee pads and a light jacket and trousers." Among them, "If you have rheumatoid arthritis, you can bring cold-proof knee pads and a light jacket and trousers" is added to the answer to this query based on the historical information of the previous query "I have rheumatoid arthritis and I need to pay attention to it in my daily life."

[0163] The user-preference-based search and question-answering process proposed in this application embodiment allows for customization of user style selections and historical memory selections. It also utilizes a historically informed answer generation model and a sparsely activated style text generation model to generate heartfelt answers and personalized answers of corresponding styles. By leveraging the user's historical information to develop a personalized style and customize personalized settings, a personalized search and question-answering experience tailored to user preferences is provided.

[0164] The above describes the search question and answer system and the search question and answer method. The following describes the related devices with reference to the accompanying drawings.

[0165] As shown in FIG11 , a structure of a client 110 provided in an embodiment of the present application includes:

[0166] The sending unit 1101 is configured to send a first query request and corresponding historical information to the cloud, where the historical information includes at least one of historical query requests and corresponding answers associated with the first query request.

[0167] The receiving unit 1102 is configured to receive a first answer from the cloud, where the content of the first answer is associated with the historical information.

[0168] The display unit 1103 is configured to present the first answer.

[0169] In this application, the first answer is generated in the cloud based on historical information from the client. The historical information can reflect the characteristics of the user using the client. The first answer generated in this way is more in line with the user's perception and is a high-quality answer that conforms to the user's personality, which can improve the matching degree between the answer and the user.

[0170] Optionally, the client 110 further includes: a processing unit 1104 for parsing the first query request to retrieve historical information from a plurality of historical query requests and corresponding answers stored in the client, wherein the similarity between the historical information and the first query request is greater than a first threshold.

[0171] Optionally, the first answer is determined by the cloud based on the second query request, and the second query request is obtained by the cloud by rewriting the first query request based on historical information.

[0172] Optionally, the first query request further includes a first style selected by the user, where the first style is used to indicate a presentation form of the first answer; the display unit 1103 is used to present the content of the first answer in the first style.

[0173] Optionally, the first style is selected by the user from multiple styles through a style button on the search question and answer interface, and multiple styles are displayed after the style button is triggered.

[0174] Optionally, the first answer includes a second style, which is a presentation form of the first answer determined by the cloud based on historical information; the display unit 1103 is used to present the content of the first answer in the second style.

[0175] Optionally, the sending unit 1101 is further used to send a clarification response to the cloud when there is a conflict between the first query request and at least one of the historical query requests, and the clarification response is used to instruct the cloud to ignore the conflicting historical query requests, or the clarification response is used to instruct the cloud not to ignore the conflicting historical query requests.

[0176] Optionally, if the clarification response is used to instruct the cloud to ignore the conflicting historical query request, the first answer does not include information associated with the conflicting historical query request.

[0177] Optionally, if the clarification response is used to instruct the cloud to not ignore conflicting historical query requests, the first answer includes conflict prompt information, where the conflict prompt information is based on information determined based on the conflicting historical query requests and the first query request.

[0178] Optionally, the processing unit 1104 is further configured to respond to the user's selection of a memory mode, determine corresponding historical information for the first query request, and present the memory mode in the form of a button on the search question and answer interface.

[0179] In the embodiment of the present application, the operations performed by each unit in the client 110 are similar to those described in the embodiments shown in Figures 3 to 10 above, and will not be repeated here.

[0180] As shown in FIG12 , the embodiment of the present application further provides another structure of the cloud device 120 including:

[0181] The receiving unit 1201 is configured to receive a first query request and corresponding historical information from a client, where the historical information includes at least one of historical query requests associated with the first query request and corresponding answers.

[0182] The processing unit 1202 is configured to generate a first answer according to the first query request and the historical information, where the content of the first answer is associated with the historical information.

[0183] The sending unit 1203 is configured to send the first answer to the client.

[0184] In this application, the cloud generates a first answer to the first query request through guidance from historical information from the client. The historical information can reflect the characteristics of the user using the client, so that the first answer generated in this way is more in line with the user's perception and can improve the matching degree between the answer and the user.

[0185] Optionally, the first query request also includes a first style selected by the user, and the first style is used to indicate the presentation form of the first answer; the processing unit 1202 is specifically used to determine the content of the first answer based on the second query request and historical information, and to integrate the first style with the content of the first answer to obtain the first answer.

[0186] Optionally, the processing unit 1202 is specifically used to determine the user's style preference based on historical information, and determine a second style based on the style preference, where the second style is used to indicate the presentation form of the first answer; determine the content of the first answer based on the second query request and historical information, and integrate the second style with the content of the first answer to obtain the first answer.

[0187] Optionally, the processing unit 1202 is further configured to determine whether there is a conflict between the first query request and a historical query request.

[0188] The sending unit 1203 is further configured to send a query clarification request to the client if there is a conflict, the query clarification request being used to indicate that the first query request conflicts with at least one of the historical query requests and to inquire whether to ignore the conflicting historical query request.

[0189] The receiving unit 1201 is further configured to receive a clarification response from the client, where the clarification response is used to indicate whether to ignore the conflicting historical query request or not to ignore the conflicting historical query request.

[0190] Optionally, processing unit 1202 is further used to delete the conflicting historical query request and the corresponding answer from the historical information before generating the first answer based on the first query request and the historical information if the clarification response is used to instruct the cloud to ignore the conflicting historical query request; accordingly, the first answer does not contain information associated with the conflicting historical query request.

[0191] Optionally, if the clarification response is used to instruct the cloud not to ignore conflicting historical query requests, then correspondingly, the first answer includes conflict prompt information, and the conflict prompt information is based on information determined by the conflicting historical query requests and the first query request.

[0192] In the embodiment of the present application, the operations performed by each unit in the cloud device 110 are similar to those described in the embodiments shown in Figures 3 to 10 above, and will not be repeated here.

[0193] In another embodiment of the present application, a computer-readable storage medium is also provided, in which computer-executable instructions are stored. When the processor of the cloud device executes the computer-executable instructions, the cloud device executes the steps executed by the cloud in Figures 3 to 10 above.

[0194] In another embodiment of the present application, a computer-readable storage medium is provided, in which computer-executable instructions are stored. When the processor of the client executes the computer-executable instructions, the client executes the steps performed by the client in Figures 3 to 10 above.

[0195] In another embodiment of the present application, a computer program product is further provided. The computer program product includes computer program code. When the computer program code is executed on a computer, the computer device executes the steps executed by the cloud or client in Figures 3 to 10 above.

[0196] In another embodiment of the present application, a chip system is also provided, which includes one or more interface circuits and one or more processors; the interface circuit and the processor are interconnected by lines; the interface circuit is used to receive signals from the memory of the terminal and send signals to the processor, and the signals include computer instructions stored in the memory; when the processor executes the computer instructions, the terminal executes the steps performed by the cloud device or client in Figures 3 to 10 above. In one possible design, the chip system may also include a memory, which is used to store program instructions and data necessary for the control device. The chip system can be composed of chips, or it can include chips and other discrete devices.

[0197] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0198] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0199] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in whole or in part through software, hardware, firmware, or any combination thereof.

[0200] When software is used to implement the integrated unit, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid state drive (SSD)).

Claims

1. A method for searching questions and answers, characterized in that: include: Sending a first query request and corresponding historical information to a cloud, wherein the historical information includes at least one of a historical query request and a corresponding answer associated with the first query request; receiving a first answer from the cloud, wherein content of the first answer is associated with the historical information; The first answer is presented.

2. The method according to claim 1, characterized in that Before sending the first query request and the corresponding historical information to the cloud, the method further includes: The first query request is parsed to retrieve the historical information from a plurality of historical query requests and corresponding answers stored in the client, wherein the similarity between the historical information and the first query request is greater than a first threshold.

3. The method according to claim 1 or 2, characterized in that: The first answer is determined by the cloud based on a second query request, and the second query request is obtained by the cloud rewriting the first query request based on the historical information.

4. The method according to any one of claims 1 to 3, characterized in that: The first query request further includes a first style selected by a user, where the first style is used to indicate a presentation form of the first answer; The presenting the first answer includes: presenting the content of the first answer in the first style.

5. The method according to claim 4, characterized in that The first style is selected by the user from a plurality of styles through a style button on the search question and answer interface, and the plurality of styles are displayed after the style button is triggered.

6. The method according to any one of claims 1 to 3, characterized in that: The first answer includes a second style, where the second style is a presentation form of the first answer determined by the cloud according to the historical information; The presenting the first answer includes: presenting the content of the first answer in the second style.

7. The method according to any one of claims 1 to 6, characterized in that: The method further comprises: If the first query request conflicts with at least one of the historical query requests, a clarification response is sent to the cloud, wherein the clarification response is used to instruct the cloud to ignore the conflicting historical query request, or the clarification response is used to instruct the cloud not to ignore the conflicting historical query request.

8. The method according to claim 7, characterized in that If the clarification response is used to instruct the cloud to ignore the conflicting historical query request, the first answer does not include information associated with the conflicting historical query request.

9. The method according to claim 7, characterized in that: If the clarification response is used to instruct the cloud not to ignore conflicting historical query requests, the first answer includes conflict prompt information, where the conflict prompt information is determined based on information of the conflicting historical query requests and the first query request.

10. The method according to any one of claims 1 to 9, characterized in that: The method further comprises: In response to the user's selection of the memory mode, the corresponding historical information is determined for the first query request, and the memory mode is presented in the form of a button on the search question and answer interface.

11. A method for searching questions and answers, characterized in that: include: receiving a first query request and corresponding historical information from a client, wherein the historical information includes at least one of historical query requests and corresponding answers associated with the first query request; generating a first answer according to the first query request and the historical information, wherein the content of the first answer is associated with the historical information; The first answer is sent to the client.

12. The method according to claim 11, characterized in that The generating a first answer according to the first query request and the historical information includes: Rewriting the first query request according to the historical information to obtain a second query request; The first answer is generated according to the second query request and the historical information.

13. The method according to claim 12, characterized in that The first query request further includes a first style selected by a user, where the first style is used to indicate a presentation form of the first answer; The generating a first answer according to the second query request and the historical information includes: The content of the first answer is determined according to the second query request and the historical information, and the first style is integrated with the content of the first answer to obtain the first answer.

14. The method according to claim 12, characterized in that The generating a first answer according to the second query request and the historical information includes: determining a user's style preference according to the historical information, and determining a second style according to the style preference, wherein the second style is used to indicate a presentation form of the first answer; The content of the first answer is determined according to the second query request and the historical information, and the second style is integrated with the content of the first answer to obtain the first answer.

15. The method according to any one of claims 11 to 14, characterized in that: The method further comprises: Determine whether the first query request conflicts with the historical query request; If there is a conflict, sending a query clarification request to the client, the query clarification request is used to indicate that the first query request conflicts with at least one of the historical query requests, and to inquire whether to ignore the conflicting historical query request; A clarification response is received from the client, where the clarification response is used to indicate to ignore the conflicting historical query request or not to ignore the conflicting historical query request.

16. The method according to claim 15, characterized in that If the clarification response is used to instruct the cloud to ignore the conflicting historical query request, before generating a first answer according to the first query request and the historical information, deleting the conflicting historical query request and the corresponding answer from the historical information; Correspondingly, the first answer does not include information associated with the conflicting historical query request.

17. The method according to claim 15, characterized in that If the clarification response is used to instruct the cloud not to ignore conflicting historical query requests, then correspondingly, the first answer includes conflict prompt information, where the conflict prompt information is determined based on information of the conflicting historical query requests and the first query request.

18. A client, characterized in that: include: a sending unit, configured to send a first query request and corresponding historical information to a cloud, wherein the historical information includes at least one of a historical query request and a corresponding answer associated with the first query request; A receiving unit, configured to receive a first answer from the cloud, wherein the content of the first answer is associated with the historical information; A display unit is used to present the first answer.

19. A cloud device, characterized in that: include: a receiving unit, configured to receive a first query request and corresponding historical information from a client, wherein the historical information includes at least one of a historical query request and a corresponding answer associated with the first query request; a processing unit, configured to generate a first answer according to the first query request and the historical information, wherein the content of the first answer is associated with the historical information; A sending unit is used to send the first answer to the client.

20. A client, characterized in that: include: A transceiver, a processor and a memory, wherein the transceiver and the processor are coupled to the memory, and the memory is used to store programs or instructions. When the programs or instructions are executed by the processor, the client executes the method according to any one of claims 1 to 10.

21. A cloud device, characterized in that: include: A communication interface, a processor and a memory, wherein the communication interface and the processor are coupled to the memory, and the memory is used to store programs or instructions. When the programs or instructions are executed by the processor, the cloud device executes the method as described in any one of claims 11 to 17.

22. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a computer device, the computer device is caused to perform the method according to any one of claims 1 to 17.

23. A computer program product, characterized in that The computer program product comprises a computer program code, which, when executed on a computer device, causes the computer device to perform the method according to any one of claims 1 to 17.

24. A chip system, characterized in that: The chip system includes one or more interface circuits and one or more processors; the interface circuit and the processor are interconnected by lines; the interface circuit is used to receive signals from the client's memory and send signals to the processor, the signals including computer instructions stored in the memory; when the processor executes the computer instructions, the client executes the method as described in any one of claims 1 to 10; or, the interface circuit is used to receive signals from the memory of the cloud device and send signals to the processor, the signals including computer instructions stored in the memory; when the processor executes the computer instructions, the cloud device executes the method as described in any one of claims 11 to 17.

25. A search question answering system, characterized in that: include: A client and a cloud device, wherein the client is used to execute the method described in any one of claims 1 to 10, and the cloud device is used to execute the method described in any one of claims 11 to 17.

Images