Question and answer method and related device

Abstract

The invention discloses a question and answer method and a related device, and the method comprises the steps: obtaining inquiry information inputted by a target object, and obtaining reply reference information matched with the inquiry information through a target agent; based on the reply reference information, generating a reply result matched with the inquiry information; wherein the reply result comprises reply data matched with at least one display type, and the display type is displayed according to a target display form determined by the target object; and obtaining an association instruction triggered by the target object according to the reply result, and generating a matched feedback result based on the association instruction. Through the mode, the accuracy of feeding back the inquiry information of the target object can be improved, and the experience feeling of the target object is improved.

Description

Technical Field

[0001] This application relates to the field of natural language processing technology, and in particular to a question-answering method and related apparatus. Background Technology

[0002] With the rapid development of informatization and intelligentization, data acquisition and processing have become increasingly important. Intelligent question-answering systems, based on natural language processing technology, are widely used in customer service, knowledge management, data retrieval, and other fields. Existing technologies mainly use keyword matching, rule engines, and simple machine learning models to understand and answer input queries. However, traditional methods have many shortcomings in terms of the accuracy of intent understanding and the flexibility of response display, failing to meet the growing demands.

[0003] Therefore, how to propose a highly accurate intelligent question-answering method has become an urgent problem to be solved. Summary of the Invention

[0004] The main technical problem addressed by this application is to provide a question-and-answer method and related apparatus that can improve the accuracy of feedback on the query information of the target object and enhance the target object's experience.

[0005] To address the aforementioned technical problems, this application provides a question-and-answer method, comprising: acquiring query information input by a target object; acquiring response reference information matching the query information using a target intelligent agent; generating a response result matching the query information based on the response reference information; wherein the response result includes response data matching at least one display type, the display type being displayed according to a target display format determined by the target object; acquiring an association instruction triggered by the target object based on the response result; and generating a matching feedback result based on the association instruction.

[0006] To address the aforementioned technical problems, a second aspect of this application provides an electronic device including a memory and a processor coupled to each other, wherein the memory stores program instructions and the processor executes the program instructions to implement the methods mentioned in the above technical solutions.

[0007] To address the aforementioned technical problems, a third aspect of this application provides a computer-readable storage medium storing program instructions executable by a processor, the program instructions being used to implement the methods mentioned in the above technical solutions.

[0008] The beneficial effects of this application are as follows: Unlike existing technologies, the question-answering method proposed in this application, after obtaining the query information from the target object, calls the target agent to obtain response reference information matching the query information, thereby improving the accuracy of understanding the query information during the question-answering interaction. Based on the response reference information, a more accurate response result is generated, and the response result includes response data matching the first display type. The first display type is displayed according to the target display format determined by the target object, improving the flexibility of data display. Furthermore, after displaying the response result, by obtaining the associated instructions triggered by the target object based on the response result, a further round of question-answering interaction is conducted, enhancing the relevance of question-answering interactions in different scenarios and helping to improve the user experience of the target object in question-answering interactions. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein:

[0010] Figure 1 This is a flowchart illustrating one implementation method of the question-and-answer method of this application;

[0011] Figure 2 yes Figure 1 The flowchart of step S101 corresponds to another embodiment;

[0012] Figure 3 yes Figure 1 The flowchart of step S103 corresponds to another embodiment;

[0013] Figure 4 This is a schematic diagram of the display interface corresponding to one embodiment of step S302;

[0014] Figure 5 yes Figure 1 The flowchart of step S103 corresponds to another embodiment;

[0015] Figure 6 This is a schematic diagram of the display interface corresponding to one embodiment of step S402;

[0016] Figure 7 yes Figure 1 The flowchart of step S103 corresponds to another embodiment;

[0017] Figure 8 This is a schematic diagram of the display interface corresponding to one embodiment of step S502;

[0018] Figure 9 yes Figure 1 The flowchart of step S103 corresponds to another embodiment;

[0019] Figure 10 yes Figure 1 The flowchart of another embodiment corresponds to step S103;

[0020] Figure 11 yes Figure 10 The flowchart of step S702 corresponds to another embodiment;

[0021] Figure 12 This is a schematic diagram of the structure of one embodiment of the electronic device of this application;

[0022] Figure 13 This is a schematic diagram of one embodiment of the computer-readable storage medium of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments, and different implementation methods can be adaptively combined. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0024] In this paper, the terms "system" and "network" are often used interchangeably. The term "and / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. Additionally, the character " / " generally indicates that the preceding and following related objects have an "or" relationship. Furthermore, "many" in this paper means two or more.

[0025] The question-answering method provided in this application is used in a question-answering system, which includes a question-answering processing end and a target intelligent agent that are coupled to each other. The execution subject corresponding to the above question-answering method is a processing unit capable of data processing. This processing unit is integrated into the question-answering processing end or exists independently of the question-answering processing end and interacts with the question-answering processing end for data processing. The question-answering processing end can be a smart device or a webpage, etc., and the smart device includes any one of a mobile phone, computer, or wearable device.

[0026] Please see Figure 1 , Figure 1 This is a flowchart illustrating one implementation of the question-and-answer method of this application, which includes:

[0027] S101: Obtain the query information input by the target object, and use the target intelligent agent to obtain response reference information that matches the query information.

[0028] In one embodiment, query information input by the target object is obtained, the query information is input into the target intelligent agent, and the target intelligent agent is used to obtain response reference information that matches the query information.

[0029] Specifically, in response to the target object inputting query information on the question-and-answer processing terminal, the query information is obtained and then input into the target intelligent agent. The target intelligent agent uses information recognition and extraction on the query information to generate response reference information.

[0030] In some implementation scenarios, the question-and-answer processing terminal includes a display interface with an information input area. In response to a target object triggering an input command in this area, the system retrieves the input query information. The target agent, possessing natural language processing capabilities, extracts at least one key piece of information from the query information—including intent, entity, and context—to generate a matching response reference.

[0031] In some implementation scenarios, the above-mentioned query information is text. Alternatively, the query information can also be text obtained by recognizing audio input from the target object.

[0032] In some implementation scenarios, the aforementioned target intelligent agent includes a natural language processing model, which can be a large language model with superior data analysis capabilities; or, the natural language processing model can be a neural network model trained using multiple training samples.

[0033] S102: Based on the response reference information, generate a response result that matches the query information; wherein, the response result includes response data that matches at least one first display type, and the first display type is displayed according to the target display format determined by the target object.

[0034] In one embodiment, reference information for the response generated by the target agent is obtained, and a response result matching the query information is generated based on the query information and the reference response information.

[0035] In some implementation scenarios, a response result is generated by combining reference response information to answer the inquiry. The response result includes response data that matches at least one first display type, and the first display type is displayed according to the target display format determined by the target object. The at least one first display type includes at least one of a chart type, a graph type, and a scene type.

[0036] In some implementation scenarios, the question-and-answer processing end includes an intelligent analysis model. After obtaining the answer reference information, the answer reference information and the question information are input into the intelligent analysis model. The intelligent analysis model then summarizes and responds to the question information based on the answer reference information, thereby generating the answer result. The aforementioned intelligent analysis model is a large language model with superior data analysis capabilities.

[0037] In a specific application scenario, the aforementioned large language model may include, but is not limited to, deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory networks (LSTM), and generative pre-trained Transformer models, etc. No specific restrictions are placed on the specific construction and deployment of the large language model. Furthermore, it should be noted that the specific construction and deployment of the intelligent analysis models mentioned in other embodiments of this application can refer to this embodiment.

[0038] In another embodiment, in response to the question-and-answer processing terminal including a target database, the target database is used to store data acquired through a specific method and / or data related to the historical interaction process of the target object. Based on the acquired reference response information, relevant information related to the reference response information is retrieved from the target database. Based on the aforementioned relevant information and the reference response information, a response result matching the query information is generated.

[0039] In some implementation scenarios, the question-and-answer processing terminal is associated with multiple devices for data collection, and the target database mentioned above includes data collected by the associated devices.

[0040] S103: Obtain the associated instructions triggered by the target object based on the response result, and generate matching feedback results based on the associated instructions.

[0041] In one embodiment, after the generated response result is displayed, in response to the target object triggering an associated instruction on the display interface based on the response result, the associated instruction is obtained and a feedback result matching the associated instruction is generated.

[0042] In some implementation scenarios, the system acquires relevant information actively input by the target user based on the displayed response, and triggers related instructions to prompt processing of that information. This relevant information is used to ask follow-up questions about parts of the generated response or to request the display of detailed information related to certain parts of the response.

[0043] In a specific application scenario, the response displayed includes relationship graph data, which contains multiple reference entities and the relationships between them. When a referencing command triggered by the target object clicking on any reference entity is received, detailed content matching the corresponding reference entity is generated and displayed on the interface. Alternatively, when a referencing command triggered by the target object clicking on any relationship is received, detailed content matching the corresponding relationship is generated, including information related to the corresponding reference entity, and this detailed content is also displayed on the interface.

[0044] The question-answering method proposed in this application, after obtaining the query information from the target object, calls the target agent to obtain response reference information matching the query information, thereby improving the accuracy of understanding the query information during the question-answering interaction. Based on the response reference information, a more accurate response result is generated, and the response result includes response data matching a first display type. The first display type displays data according to the target display format determined by the target object, improving the flexibility of data display. Furthermore, after displaying the response result, by obtaining the associated instructions triggered by the target object based on the response result, a further round of question-answering interaction is conducted, enhancing the relevance of question-answering interactions in different scenarios and helping to improve the target object's experience in question-answering interactions.

[0045] Please see Figure 2 , Figure 2 yes Figure 1 The flowchart of step S101 corresponds to another embodiment. Specifically, the implementation process of step S101 includes:

[0046] S201: Input the query information into the target agent and use the target agent to obtain the identification element information that matches the query information.

[0047] In one embodiment, the query information input by the target object is input into the target intelligent agent, and the target intelligent agent is used to perform element recognition on the query information to obtain the corresponding recognition element information.

[0048] Specifically, after the query information is input into the target intelligent agent, the target intelligent agent is used to identify and extract the element categories such as intent element category, entity element category and scene element category in the query information, and the extracted information is used as the identification element information.

[0049] S202: Determine whether the identified element information meets the preset response conditions.

[0050] In one embodiment, after obtaining the identification element information output by the target intelligent agent, it is determined whether the identification element information meets preset response conditions. These preset response conditions are related to the completeness of all identification element information.

[0051] Specifically, based on all the acquired identification element information, it is determined whether any necessary information for generating a response is missing. If missing, the identification element information is deemed not to meet the preset response conditions. If not missing, the identification element information is deemed to meet the preset response conditions.

[0052] In some implementation scenarios, the question-and-answer processing end includes an intelligent analysis model and a target database. After obtaining the identification element information output by the target intelligent agent, the identification element information is input into the intelligent analysis model of the question-and-answer processing end. The intelligent analysis model, combined with the target database, determines whether the identification element information is complete. If the identification element information is complete, it is determined that the preset response conditions are met; if it is incomplete, it is determined that the preset response conditions are not met.

[0053] S203: In response to the fact that the identified element information meets the preset response conditions, the identified element information is used as the response reference information.

[0054] In one embodiment, when the identified element information meets the preset response conditions, the identified element information is used as response reference information so that a response result can be generated subsequently based on the response reference information.

[0055] S204: In response to the fact that the identified element information does not meet the preset response conditions, the target agent obtains supplementary information matching the query information from the reference database, and obtains response reference information based on the identified element information and the supplementary information.

[0056] In one embodiment, the target agent includes a reference database. When the identified element information does not meet the preset response conditions, to improve the accuracy of the generated response, the target agent uses the reference database to search for supplementary information that matches the query information. The identified element information and the supplementary information are used as reference information for the response.

[0057] In one embodiment, when the identified element information does not meet the preset response conditions, the target agent retrieves supplementary information matching the query information from a reference database and uses this supplementary information to supplement the already acquired identified element information. Further, it is determined whether the identified element information meets the preset response conditions. If it does, the supplemented identified element information is used as the response reference information. If it does not, the missing information in the generated response result is determined based on the supplemented identified element information. A prompt message is generated based on the missing information to prompt the target object to supplement the missing information, and this prompt message is displayed on the display interface. In response to obtaining the detailed information input by the target object based on the prompt message, the detailed information and the supplemented identified element information are used as the response reference information.

[0058] The above scheme, by combining the target agent and the reference database, improves the completeness of the response reference information and helps to improve the accuracy of the generated response results.

[0059] Please see Figure 3 , Figure 3 yes Figure 1 The flowchart of step S103 corresponds to another embodiment. Specifically, the implementation process of step S103 includes:

[0060] S301: Based on the response reference information, generate chart data that matches multiple candidate display formats.

[0061] In one implementation, the response reference information is analyzed, and chart data matching multiple candidate display formats is generated.

[0062] In a specific application scenario, the candidate display formats for icon data include at least some of the following: bar chart, column chart, line chart, and pie chart.

[0063] In another embodiment, in response to the intention information included in the response reference information to generate chart data, the response reference information is analyzed and chart data matching multiple candidate display formats is generated.

[0064] S302: Display the candidate display formats on the display interface, obtain the target display format selected by the target object from all candidate display formats, and display the chart data corresponding to the target display format.

[0065] In one implementation, please refer to Figure 4 , Figure 4 This is a schematic diagram of the display interface corresponding to one embodiment of step S302. Candidate display formats are displayed on the display interface, the target object is selected from all candidates in the target display format, and the chart data corresponding to the target display format is displayed on the display interface, improving the flexibility of chart data display.

[0066] In some implementation scenarios, candidate identifiers are generated that match each candidate display format, i.e. Figure 4 The "bar chart," "pie chart," "line chart," and "column chart" shown herein display candidate icons on the interface. In response to a target object clicking on any candidate icon, the candidate display format corresponding to the clicked icon is used as the target display format, and the chart data corresponding to the target display format is displayed.

[0067] In some implementation scenarios, after generating chart data that matches multiple candidate display formats, a default display format is randomly determined from these candidates. The chart data corresponding to the default display format, along with candidate icons for all candidate display formats, is then displayed on the interface. In response to a target object clicking a candidate icon on the display interface, the candidate display format corresponding to the clicked icon is chosen as the target display format, and the currently displayed chart data is switched to the chart data corresponding to the target display format. For example, ... Figure 4 As shown, after generating chart data that matches multiple candidate display formats, the candidate display format corresponding to "bar chart" is used as the default display format, and the corresponding chart data is displayed on the display interface; in response to the target object clicking "line chart", the chart data corresponding to the currently displayed "bar chart" is switched to the chart data corresponding to "line chart".

[0068] Please see Figure 5 , Figure 5 yes Figure 1 The flowchart of step S103 corresponds to another embodiment. Specifically, the implementation process of step S103 includes:

[0069] S401: Based on the response reference information, generate relationship graph data that matches multiple candidate display formats; wherein, the relationship graph data includes multiple reference graph elements, which are reference entities or the relationships between different reference entities, and the candidate display formats are determined based on the degree of association between different reference entities.

[0070] In one embodiment, the response reference information is analyzed, and relationship graph data matching multiple candidate display formats is generated. The relationship graph data includes multiple reference graph elements, which are reference entities or relationships between different reference entities. Furthermore, different candidate display formats are determined based on the degree of association between different reference entities.

[0071] Specifically, after analyzing the response reference information, multiple reference entities are identified, including a core entity and related entities connected to the core entity. The correlation degree between each related entity and the core entity is obtained, and all related entities are divided into multiple entity sets based on the correlation degree. Candidate display formats corresponding to each entity set are determined, and relationship graph data matching the corresponding candidate display formats is generated based on the core entity and each entity set. In the relationship graph data, core entities and related entities are connected by connecting lines, and these connecting lines are matched with relationship identifiers. These relationship identifiers represent the association between the core entity and its corresponding related entity.

[0072] In some implementation scenarios, within relationship graph data, certain associated entities are matched with key identifiers and key annotations. The key identifiers are determined based on the degree of association and / or relationship between the corresponding associated entity and the core entity, while the key annotations explain the reason for setting the key identifiers. For example, associated entities with an association degree greater than a target threshold are assigned key identifiers, and key annotations are added next to these key identifiers to illustrate the high degree of association between the corresponding associated entity and the core entity.

[0073] In a specific application scenario, based on the degree of association between each associated entity and the core entity, all associated entities are divided into a first entity set, a second entity set, and a third entity set. The degree of association between any associated entity in the first entity set and the core entity is greater than or equal to a first threshold. The degree of association between any associated entity in the second entity set and the core entity is greater than or equal to a second threshold and less than the first threshold, where the second threshold is less than the first threshold. Similarly, the degree of association between any associated entity in the third entity set and the core entity is greater than or equal to a third threshold and less than the second threshold, where the third threshold is less than the second threshold.

[0074] S402: Display the candidate display formats on the display interface, obtain the target display format selected by the target object from all candidate display formats, and display the relationship graph data corresponding to the target display format.

[0075] In one implementation, a candidate identifier matching each candidate display format is generated, and the candidate identifiers are displayed on the display interface. In response to a target object clicking any candidate identifier, the candidate display format corresponding to the clicked candidate identifier is used as the target display format, and the relationship graph data corresponding to the target display format is displayed, improving the flexibility of displaying the relationship graph data.

[0076] For a specific application scenario, please refer to Figure 6 , Figure 6 This is a schematic diagram of the display interface corresponding to one embodiment of step S402. Based on the degree of association, relationship graph data corresponding to different candidate display formats are determined, namely "first-degree relationship graph data," "second-degree relationship graph data," and "third-degree relationship graph data," with "first-degree relationship graph data" displayed by default. In the "first-degree relationship graph data," the degree of association between any associated entity and the core entity is greater than the degree of association between any associated entity and the core entity in the "second-degree relationship graph data" or "third-degree relationship graph data." In response to the target object clicking on "second-degree relationship graph data" or "third-degree relationship graph data" on the display interface, the corresponding relationship graph data is switched to be displayed on the display interface.

[0077] Please see Figure 7 , Figure 7 yes Figure 1 The flowchart of step S103 corresponds to another embodiment. Specifically, the implementation process of step S103 includes:

[0078] S501: Based on the response reference information, generate scene data that matches multiple candidate display formats; wherein, the scene data matched by each candidate display format includes at least one labeled location information.

[0079] In one embodiment, the response reference information is analyzed and scene data matching multiple candidate display formats is generated, wherein the scene data matching each candidate display format includes at least one annotation location information.

[0080] Specifically, after analyzing the response reference information, a target map matching the response reference information and at least one labeled location information in the target map are generated. Based on the target map and labeled location information, scene data matching different candidate display formats is generated. The display method for different labeled location information in the scene data differs under different candidate display formats.

[0081] In some implementation scenarios, for the target map and the labeled location information within the target map, the various labeled location information are connected to form scene data that matches one candidate display format. Conversely, for the target map and the labeled location information within the target map, each labeled location information is labeled individually to form scene data that matches another candidate display format.

[0082] In a specific application scenario, when the target object inputs a query requesting the generation of a movement path matching a certain vehicle, matching response reference information is obtained through the aforementioned implementation method. Based on the response reference information, a target map matching the vehicle and at least one marked location information in the target map are generated. The marked location information represents areas where the vehicle has passed through or stayed for more than a preset time. According to the vehicle's travel sequence, the marked location information on the target map is sequentially connected to obtain scene data matching one candidate display format. The scene data in this candidate display format can intuitively show the vehicle's travel process and path. Furthermore, the marked location information on the target map is marked with preset identifiers to obtain scene data matching another candidate display format. The scene data in this candidate display format can intuitively show the areas traversed by the vehicle.

[0083] S502: Display the candidate display formats on the display interface, obtain the target display format selected by the target object from all candidate display formats, and display the scene data corresponding to the target display format.

[0084] In one implementation, a candidate identifier matching each candidate display format is generated, and the candidate identifiers are displayed on the display interface. In response to a target object clicking any candidate identifier, the candidate display format corresponding to the clicked candidate identifier is taken as the target display format, and the scene data corresponding to the target display format is displayed.

[0085] For a specific application scenario, please refer to Figure 8 , Figure 8 This is a schematic diagram of the display interface corresponding to one embodiment of step S502. Based on the target map matching the response reference information and the marked location information, scene data corresponding to the "location marker map" and scene data corresponding to the "path map" are generated, and the scene data corresponding to the "location marker map" is displayed by default. In response to the target object clicking the "path map" in the display interface, the scene data currently displayed on the interface is switched.

[0086] Alternatively, in other embodiments, Figure 1 The response results generated in step S102 include response data matching multiple first display types, and some of these first display types are displayed according to the target display format determined by the target object. For example, when the generated response results include chart data matching a chart type, relationship graph data matching a graph type, and scene data matching a scene type, the chart type corresponds to chart data matching multiple candidate display formats, and the corresponding chart data is displayed according to the target display format determined by the target object. The graph type and scene type only correspond to relationship graph data and scene data matching the default display format.

[0087] Please see Figure 9 , Figure 9 yes Figure 1 The flowchart of step S103 corresponds to another embodiment. Specifically, the implementation process of step S103 includes:

[0088] S601: Based on the response results, generate multiple predicted questions related to the response results and display the predicted questions on the display interface.

[0089] In one embodiment, multiple predictive question messages are generated for the generated response results, and the generated predictive question messages are displayed on the display interface.

[0090] In some implementation scenarios, to predict the follow-up questions that the target audience will ask based on the response, an intelligent analysis model is used to generate and display multiple predicted questions based on the response information. This allows the target audience to ask follow-up questions about the generated response by clicking on any of the predicted questions, thus improving the efficiency of question-and-answer interaction.

[0091] In a specific application scenario, such as Figure 4 As shown, while displaying the chart data corresponding to the target display format on the interface, the generated predicted question information a, predicted question information b, and predicted question information c are displayed below the chart data. It should be noted that the actual number of predicted question information items generated can be set according to the actual scenario and requirements.

[0092] S602: In response to the association instruction triggered by the target object selecting any predicted question information, based on the association instruction, the target agent obtains feedback reference information that matches the predicted question information corresponding to the association instruction.

[0093] In one embodiment, when the target object clicks on any predicted question on the display interface, an association instruction for following up on the answer is triggered. The predicted question and answer corresponding to the association instruction are input into the target agent, which then performs element recognition on the predicted question and answer to obtain the recognized element information corresponding to the predicted question. The recognized element information corresponding to the predicted question is used as feedback reference information. The specific process for obtaining the feedback reference information can refer to the process for obtaining the answer reference information in the corresponding embodiment described above, and will not be elaborated here.

[0094] In another embodiment, after obtaining the identification element information corresponding to the predicted question information using the target agent, it is determined whether the identification element information corresponding to the predicted question information meets the preset response conditions. If it does, the identification element information corresponding to the predicted question information is used as feedback reference information. Otherwise, the target agent obtains supplementary feedback information from the reference database, and uses the supplementary feedback information to supplement the identification element information corresponding to the predicted question information to obtain feedback reference information. Specific implementation can refer to the corresponding embodiments described above.

[0095] S603: Based on the feedback reference information, generate feedback results that match the associated instructions.

[0096] In one embodiment, the predicted question information corresponding to the associated instruction is responded to based on the obtained feedback reference information and response results, generating a matching feedback result. The display type of the feedback result may be consistent with the first display type mentioned in the corresponding embodiments above, or the display type of the feedback result may be different from the first display type mentioned in the corresponding embodiments above.

[0097] In some implementation scenarios, feedback reference information is input into the intelligent analysis model so that the intelligent analysis model can combine the feedback reference information and the response results generated in the previous round to respond to the predicted question information corresponding to the associated instruction and generate a matching feedback result.

[0098] The above solution improves the efficiency of interaction between the target audience and the question-and-answer processing terminal and enhances the target audience's experience by generating multiple predictive question information related to the generated response results, thereby predicting the follow-up questions that the target audience may input based on the response results.

[0099] Please see Figure 10 , Figure 10 yes Figure 1 The flowchart of step S103 corresponds to another embodiment. Specifically, the implementation process of step S103 includes:

[0100] S701: Obtain the associated information input by the target object based on a portion of the response result, and trigger the associated instruction based on the associated information.

[0101] In one embodiment, the relevant information actively input by the target object based on the displayed response result is obtained, and a relevant instruction for prompting processing of the relevant information is triggered based on the relevant information.

[0102] In some implementation scenarios, after displaying the generated response, in response to obtaining the related information entered by the target object on the display interface based on the response, a related instruction is triggered based on the related information. This related information can be either the question text entered by the target object regarding a portion of the response; or, in response to the target object entering an audio question regarding a portion of the response, the audio question is subjected to text recognition, and the recognized text is used as the related information.

[0103] In another embodiment, after displaying the generated response results, the selection command triggered by the target object on a portion of the response results and the question content entered by the target object are obtained. The selection command and the question content are used as association information, and the association command is triggered according to the association information.

[0104] In some implementation scenarios, the selection command is triggered when the target object selects a portion of the response results displayed on the interface; the aforementioned question content can be the question text entered by the target object, or it can be obtained by text recognition based on the question audio entered by the target object.

[0105] S702: Based on the associated information and associated instructions, generate feedback results that match the second display type.

[0106] In one embodiment, a feedback result matching the second display type is generated based on the associated information and associated instructions. The second display type may be the same as the first display type mentioned in the corresponding embodiments above, or it may be different from the first display type mentioned in the corresponding embodiments above.

[0107] In some implementation scenarios, the second display type includes at least one of chart type, graph type, and scene type. Alternatively, the second display type can also be other display types related to the associated information; for example, the second display type can also be at least one of multimodal data types, such as plain text type, audio type, image type, and video type. When the second display type corresponding to the generated feedback result differs from the first display type, switching between different question-and-answer scenarios is achieved, improving the flexibility of question-and-answer interaction.

[0108] Please see Figure 11 , Figure 11 yes Figure 10 The flowchart of step S702 corresponds to another embodiment. Specifically, the associated information includes target location information, and the associated instruction is used to display the collected data related to the target location information. The implementation process of step S702 includes:

[0109] S801: Obtain at least one candidate device that matches the target location information.

[0110] In one embodiment, the associated information input in response to the target object includes at least one target location information, and the associated instruction is used to call the collected data corresponding to the target location information to obtain at least one candidate device that matches the target location information.

[0111] In some implementation scenarios, the response displayed on the current interface includes scenario data matching the scenario type. This scenario data includes a target map and marker locations on the target map. When the target object clicks on any marker location, the click is used as associated information, and the clicked marker location is used as the target location information. At least one candidate device matching the target location information is obtained.

[0112] In some implementation scenarios, the response displayed on the current interface includes scenario data matching the scenario type. This scenario data includes the target map and marked location information within the target map. The system retrieves other location information besides the marked location information clicked by the target object, treating the clicked location information as associated information and as the target location information. It also retrieves at least one candidate device matching the target location information.

[0113] In some implementation scenarios, the response results displayed on the current interface include relationship graph data matching the graph type. This relationship graph data includes multiple reference entities. When the target object clicks on any reference entity, at least one candidate location information matching the clicked reference entity is displayed on the interface. In response to the target object selecting target location information from all candidate location information, at least one candidate device matching the target location information is obtained.

[0114] In some implementation scenarios, the response results displayed on the current interface include chart data matching the chart type, scene data matching the scene type, and relationship graph data matching the graph type. The system retrieves the associated information input by the target object based on the response results. This associated information includes the question text input by the target object or the recognized text obtained from audio recognition of the question input by the target object. The system then determines the target location information within the associated information and retrieves at least one candidate device that matches the target location information.

[0115] In a specific application scenario, the aforementioned candidate devices are video stream acquisition devices. When the response result is a statistical chart of vehicles passing through location A, and the target object's input is "retrieve the video stream acquired by a video stream acquisition device near location A", then location A is taken as the target location information, and at least one candidate device matching the target location information is obtained.

[0116] In one embodiment, after obtaining at least one candidate device that matches the target location information using the general above-described embodiments, all candidate devices that match the target location information are displayed on the display interface.

[0117] S802: Determine the target device from all candidate devices and obtain the target acquisition data collected by the target device that matches the associated command.

[0118] In one embodiment, in response to the target object clicking on any candidate device, the candidate device clicked by the target object is taken as the target device, and target acquisition data matching the association instruction is determined from the original acquisition data collected from the target device according to the associated content and associated instructions.

[0119] In some implementation scenarios, the associated information in response to the target object input includes a reference entity, and the target acquisition data related to the reference entity is extracted from the raw acquisition data collected from the target device.

[0120] In some implementation scenarios, in response to the target object clicking on any candidate device and obtaining the target time interval entered by the target object on the display interface, the candidate device clicked by the target object is taken as the target device, and the original data collected by the target device within the target time interval is taken as the target collected data.

[0121] In a specific application scenario, if the reference entity is a vehicle, then after determining the target device, the target data including the reference entity is extracted from the raw data collected by the target device.

[0122] In one embodiment, in response to obtaining the description information input by the target object, target acquisition data matching the description information is extracted from the raw acquisition data collected from each candidate device.

[0123] In a specific application scenario, to acquire video capture data of a vehicle at a target location, the system obtains descriptive information about the vehicle input by the target user on a display interface. This descriptive information includes descriptive text or audio, and includes at least one of the following: time information, vehicle appearance features, vehicle model, and license plate number. Based on this descriptive information, the system analyzes the raw acquisition data collected by each candidate device to extract the target acquisition data including the aforementioned vehicle from the raw acquisition data.

[0124] S803: Displays the target data as feedback results on the display interface.

[0125] In one embodiment, the acquired target data is used as a feedback result, and the feedback result is displayed on the display interface.

[0126] The above solution displays the response results on the interface and then displays candidate devices that match the target location information based on the association command triggered by the target object. This improves the flexibility of the target object in scheduling the target data during the interaction with the question and answer processing terminal and helps to improve the processing efficiency of the target data.

[0127] In one embodiment, in response to obtaining the feedback result through the above-described corresponding implementation method, the feedback result is used as the reply result, and the process returns to step S103, that is, obtaining the association instruction triggered by the target object based on the latest displayed reply result, and conducting a new round of question-and-answer interaction based on the association instruction to generate a matching feedback result.

[0128] Please see Figure 12 , Figure 12This is a schematic diagram of one embodiment of the electronic device of this application. The electronic device includes a memory 10 and a processor 20 coupled to each other. The memory 10 stores program instructions, and the processor 20 executes the program instructions to implement the methods mentioned in any of the above embodiments. Specifically, the electronic device includes, but is not limited to, desktop computers, laptops, tablets, servers, etc., and is not limited thereto. In addition, the processor 20 may also be called a CPU (Center Processing Unit). The processor 20 may be an integrated circuit chip with signal processing capabilities. The processor 20 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor may be a microprocessor or any conventional processor. In addition, the processor 20 may be implemented by integrated circuit chips.

[0129] Please see Figure 13 , Figure 13 This is a schematic diagram of one embodiment of the computer-readable storage medium of this application. The storage medium 30 stores program instructions 40 that can be executed by a processor. When the program instructions 40 are executed by the processor, they implement the methods mentioned in any of the above embodiments.

[0130] In the several embodiments provided in this application, it should be understood that the disclosed methods and apparatus can be implemented in other ways. For example, the apparatus implementations described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0131] The units described as separate components may or may not be physically separate. 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 the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0132] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0133] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute all or part of the steps of the methods of various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0134] The above description is merely an embodiment of this application and does not limit the scope of protection of this application. Any equivalent structural or procedural transformations made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A question-and-answer method, characterized in that, include: Obtain the query information input by the target object, and use the target intelligent agent to obtain response reference information that matches the query information; Based on the response reference information, a response result matching the query information is generated; wherein, the response result includes response data matching at least one first display type, and the first display type is displayed according to the target display format determined by the target object; Obtain the associated instructions triggered by the target object based on the response result, and generate a matching feedback result based on the associated instructions.

2. The method according to claim 1, characterized in that, The step of obtaining the query information input by the target object and using the target intelligent agent to obtain response reference information matching the query information includes: The query information is input into the target intelligent agent, and the target intelligent agent is used to obtain the identification element information that matches the query information; Determine whether the identified element information meets the preset response conditions; In response to the identification element information satisfying the preset response conditions, the identification element information is used as the response reference information; In response to the identification element information not meeting the preset response conditions, the target agent obtains supplementary information matching the query information from the reference database, and obtains the response reference information based on the identification element information and the supplementary information.

3. The method according to claim 1, characterized in that, The first display type includes a chart type, the response data includes chart data matching the chart type, and the step of generating a response result matching the query information based on the response reference information includes: Based on the response reference information, generate the chart data that matches multiple candidate display formats; The candidate display formats are displayed on the display interface, the target display format selected by the target object from all the candidate display formats is obtained, and the chart data corresponding to the target display format is displayed.

4. The method according to claim 1, characterized in that, The first display type includes a graph type, the response data includes relationship graph data matching the graph type, and the step of generating a response result matching the query information based on the response reference information includes: Based on the response reference information, relationship graph data matching multiple candidate display formats is generated; wherein, the relationship graph data includes multiple reference graph elements, the reference graph elements are reference entities or the association between different reference entities, and the candidate display format is determined based on the degree of association between different reference entities; The candidate display formats are displayed on the display interface, the target display format selected by the target object from all the candidate display formats is obtained, and the relationship graph data corresponding to the target display format is displayed.

5. The method according to claim 1, characterized in that, The first display type includes a scenario type, the response data includes scenario data matching the scenario type, and the step of generating a response result matching the inquiry information based on the response reference information includes: Based on the response reference information, scene data matching multiple candidate display formats is generated; wherein, the scene data matching each candidate display format includes at least one annotation location information; The candidate display formats are displayed on the display interface, the target display format selected by the target object from all the candidate display formats is obtained, and the scene data corresponding to the target display format is displayed.

6. The method according to any one of claims 1-5, characterized in that, The step of obtaining the associated instruction triggered by the target object based on the response result, and generating a matching feedback result based on the associated instruction, includes: Based on the response result, multiple predicted questions related to the response result are generated, and the predicted questions are displayed on the display interface. In response to the association instruction triggered by the target object selecting any of the predicted question information, the target agent obtains feedback reference information matching the predicted question information corresponding to the association instruction based on the association instruction. Based on the feedback reference information, a feedback result matching the associated instruction is generated.

7. The method according to any one of claims 1-5, characterized in that, The step of obtaining the associated instruction triggered by the target object based on the response result, and generating a matching feedback result based on the associated instruction, includes: Obtain the associated information input by the target object based on a portion of the response result, and trigger the associated instruction based on the associated information; Based on the associated information and the associated instructions, a feedback result matching the second display type is generated.

8. The method according to claim 7, characterized in that, The associated information includes target location information, and the associated instruction is used to display collected data related to the target location information. The step of generating the feedback result matching the second display type based on the associated information and the associated instruction includes: Obtain at least one candidate device that matches the target location information; From all the candidate devices, determine the target device and obtain the target acquisition data collected by the target device that matches the associated instruction; The target data collected is displayed as the feedback result on the display interface.

9. An electronic device, characterized in that, The method includes a memory and a processor coupled to each other, the memory storing program instructions, and the processor executing the program instructions to implement the method of any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The system stores program instructions that can be executed by a processor, the program instructions being used to implement the method of any one of claims 1-8.

Images