Query method, device and equipment and readable storage medium

Abstract

The application discloses a query method, device and equipment and a readable storage medium. The method comprises the following steps: receiving a query request, extracting a keyword of query data corresponding to the query request, and generating a first key data group; generating a second key data group based on a target matching result and the first key data group, wherein the target matching result is a matching result with an accuracy coefficient value greater than an average accuracy coefficient value in a plurality of matching results determined based on a historical interaction data set, and the historical interaction data set comprises historical query data and a plurality of matching results corresponding to the historical query data; and responding to the query request based on the second key data group. In this way, the accuracy of the search result can be improved by generating the second key data group by combining the matching result with a high accuracy coefficient value in the historical interaction data set.

Description

Technical Field

[0001] This application belongs to the field of communication technology, and specifically relates to a query method, apparatus, device and readable storage medium. Background Technology

[0002] Artificial intelligence (AI) information retrieval is a technological means that uses artificial intelligence to quickly and accurately find and present content that meets user needs from massive amounts of information. Currently, AI-based search assistance systems often return a large number of irrelevant results when processing complex queries, requiring users to spend a lot of time filtering information. The lack of a deep understanding of user intent leads to low accuracy in search results. Summary of the Invention

[0003] This application provides a query method, apparatus, device, and readable storage medium to solve the problem of low accuracy of search results in current search assistance systems.

[0004] Firstly, embodiments of this application provide a query method, including: Receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group; Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group.

[0005] Optionally, the historical interaction dataset further includes historical interaction behavior data, and after receiving the query request, the method further includes: Based on the historical interaction data, accuracy coefficient values ​​are generated for the multiple matching results.

[0006] Optionally, the method further includes: Based on the historical interaction behavior data, the interaction behavior data of the multiple matching results are weighted and summed to obtain multiple accuracy coefficient values ​​corresponding to the multiple matching results; Based on the multiple accuracy coefficient values, the average accuracy coefficient value is obtained.

[0007] Optionally, the historical interaction data includes at least one of the following: number of clicks, dwell time, scroll depth, and feedback rating.

[0008] Optionally, the step of extracting keywords from the query data corresponding to the query request to generate a first key data group includes: Obtain the representation of the queried data; Based on the expression form of the query data, keywords of the query data are extracted, and the keywords include at least one of entity words, verbs and adjectives; Based on the keywords in the query data, the first key data group is generated.

[0009] Optionally, responding to the query request based on the second key data group includes: Based on the second key data group, obtain the target data in the target knowledge base that corresponds to the second key data group; Output the target data.

[0010] Secondly, embodiments of this application provide a query device, including: Receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group; Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group.

[0011] Optionally, the historical interaction dataset further includes historical interaction behavior data, and the device further includes: The third generation module is used to generate accuracy coefficient values ​​for the multiple matching results based on the historical interaction behavior data.

[0012] Thirdly, embodiments of this application provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the method described in the first aspect.

[0013] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described in the first aspect.

[0014] In this embodiment, a query request is received, keywords corresponding to the query data are extracted, and a first key data group is generated. Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is a matching result among multiple matching results determined based on a historical interaction dataset whose accuracy coefficient value is greater than the average accuracy coefficient value. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group. Thus, by combining matching results with higher accuracy coefficient values ​​from the historical interaction dataset to generate the second key data group, the accuracy of search results can be improved. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the query method provided in an embodiment of this application; Figure 2 This is a flowchart illustrating the query method provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the query device provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. 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.

[0017] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and are not used to describe a specified order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same class, not limited in number; for example, a first object can be one or more. Furthermore, in the specification and claims, "and" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0018] For ease of understanding, the following description, in conjunction with the accompanying drawings, further explains the query method, apparatus, device, and readable storage medium proposed in this application.

[0019] Please see Figure 1 , Figure 1A flowchart illustrating the query method provided in this application embodiment is shown in the figure. The method includes: Step 101: Receive a query request, extract keywords from the query data corresponding to the query request, and generate the first key data group.

[0020] The query request includes query data, and keywords of the query text corresponding to the query data are extracted to generate a first key data group.

[0021] It should be understood that extracting keywords from the query data may include extracting the query data corresponding to the query request from different dimensions, and combining the extracted keywords to generate a first key data set. In specific implementation, semantic analysis can be performed on the query request, and after analysis, keywords corresponding to the query data of the query request can be extracted.

[0022] The query data may include text, tables, voice, images, and videos, etc., and this embodiment does not limit the types of data.

[0023] Step 102: Generate a second key data group based on the target matching result and the first key data group.

[0024] It should be understood that the target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data.

[0025] The aforementioned accuracy coefficient value is used to evaluate the accuracy of the matching results. In an optional embodiment, one query data can correspond to multiple query results (i.e., matching results), and for each query result, an accuracy coefficient value can be generated using a dataset of user interaction behaviors related to that matching result.

[0026] In an optional embodiment, the calculation process for the accuracy coefficient value is as follows: Based on the user interaction behavior dataset, extract the first... Interaction behavior data Where m is the number of user interaction actions, the th... The data for the first interaction behavior corresponds to the user's first... Matching data group during the second query , No. Secondary interaction behavior data includes user browsing The system provides feedback information for each matching result, and each matching result corresponds to an accuracy coefficient value. ; Based on the user interaction behavior dataset, the first In the data of the interaction behavior, the first The number of clicks for each matching result is marked as , , will the The dwell time of each matching result is marked as , will the The scroll depth of each matching result is marked as , will the The feedback score for each matching result is tagged as follows: , will the In the data of this interaction behavior, the total time spent by the user browsing all matching results is marked as ; Accuracy coefficient value The calculation method is as follows: ; It should be understood that in this formula, Indicates the first The click-through rate of each matching result. This indicates the weight based on the percentage of clicks. Indicates the first The proportion of time each matching result is displayed out of the total time. This indicates the weighting based on the percentage of time spent in the room. Indicates the first The browsing speed of each matching result This indicates the weight given to browsing speed. This indicates the weight given to the feedback rating. , , and All are constants, and 1, Indicates according to , , and Weights, calculated to obtain the first In the data of the interaction behavior, the first The accuracy coefficient of each matching result .

[0027] In this embodiment, by calculating the accuracy coefficient value of the matching results, the accuracy and coverage of the search are improved, the quality of the search results is optimized, and users can obtain the information they need more quickly.

[0028] In an optional embodiment, the average accuracy coefficient value The calculation process is as follows: According to the Secondary interaction behavior data, The accuracy coefficients corresponding to each matching result are denoted as follows: , to Indicates the first to the second The accuracy coefficient corresponding to each matching result; ; It should be understood that in this formula, Indicates the first The average accuracy coefficient of all matching results in the data of each interaction behavior. .

[0029] Thus, after calculating the accuracy coefficient and average accuracy coefficient of all matching results, the accuracy coefficient of a particular matching result is... Less than or equal to the average accuracy value When the accuracy coefficient of the matching result is low, it indicates that the reference value of the matching result is low; Greater than the average accuracy value When the value of the matching result is high, it indicates that the matching result is of high reference value, and the matching result is determined as the target matching result.

[0030] The historical interaction dataset corresponds to the user making the current query request; that is, one user corresponds to one historical interaction dataset. One query dataset can correspond to multiple query results, and the accuracy of each query result's matching is determined based on the user's interaction behavior. Specifically, the accuracy of each matching result can be analyzed, and matching results with accuracy coefficients above the average value are identified as target matching results.

[0031] In an optional embodiment, user interaction data may include at least one of click count, dwell time, scroll depth, and feedback rating.

[0032] For example: after receiving a user's query request, if the number of clicks on a certain matching result generated by the user in response to the query request is greater than the average number of clicks on all matching results generated by the user in response to the query request, then that matching result is determined as the target matching result.

[0033] For example, after receiving a user's query request, if the dwell time of a certain matching result generated by the user in response to the query request is greater than the average dwell time of all matching results generated by the user in response to the query request, then the matching result is determined as the target matching result.

[0034] For example, after receiving a user's query request, if the scroll depth of a certain matching result generated by the user for the query request is greater than the average scroll depth of all matching results generated by the user for the query request, then that matching result is determined as the target matching result.

[0035] Step 103: Respond to the query request based on the second key data group.

[0036] In this way, by combining the matching results with high accuracy coefficients from the historical interaction dataset to generate a second key data set, the accuracy of search results can be improved.

[0037] Optionally, the historical interaction dataset further includes historical interaction behavior data, and after receiving the query request, the method further includes: Based on the historical interaction data, accuracy coefficient values ​​are generated for the multiple matching results.

[0038] It should be understood that the historical interaction dataset includes historical interaction behavior data, which is interaction behavior data generated based on the user's interaction behavior with the query results.

[0039] In the specific implementation, after receiving the query request in step 101, the accuracy coefficient value of multiple matching results corresponding to the historical query data is determined based on the historical interaction behavior data of the user corresponding to the current query request.

[0040] In this embodiment of the application, the accuracy coefficient values ​​of multiple matching results are generated by using historical interaction behavior data, which can improve the accuracy of the matching results.

[0041] The historical interaction data includes at least one of the following: number of clicks, dwell time, scrolling depth, and feedback rating.

[0042] Optionally, the method further includes: Based on the historical interaction behavior data, the interaction behavior data of the multiple matching results are weighted and summed to obtain multiple accuracy coefficient values ​​corresponding to the multiple matching results; Based on the multiple accuracy coefficient values, the average accuracy coefficient value is obtained.

[0043] It should be understood that multiple accuracy coefficient values ​​are generated by weighted summation of the interaction behavior data corresponding to multiple matching results. It should be noted that each matching result corresponds to one accuracy coefficient value. Thus, an average accuracy coefficient value is generated based on these multiple accuracy coefficient values.

[0044] In an optional embodiment, when the interaction behavior data includes click count, dwell time, scroll depth, and feedback rating, a weighted summation is performed on the interaction behavior data corresponding to multiple matching results. Specifically, this may include assigning weight values ​​to click count, dwell time, scroll depth, and feedback rating respectively, and summing the four influencing factors of click count, dwell time, scroll depth, and feedback rating for each matching result to obtain the accurate coefficient value for each matching result. The sum of the weights of click count, dwell time, scroll depth, and feedback rating is 1.

[0045] In this embodiment, the accuracy coefficient value of each matching result is obtained based on the interaction behavior data, which can improve the accuracy and coverage of the search and optimize the quality of the search results.

[0046] Optionally, the step of extracting keywords from the query data corresponding to the query request to generate a first key data group includes: Obtain the representation of the queried data; Based on the expression form of the query data, keywords of the query data are extracted, and the keywords include at least one of entity words, verbs and adjectives; Based on the keywords in the query data, the first key data group is generated.

[0047] It should be understood that the query data can be presented in any of the following formats: text, table, audio, image, or video.

[0048] When the query data is expressed in text form, the BERT tool can be used to segment the query text, identify entity words, verbs and adjectives, and form the first key data group. Entity words can include nouns of people, nouns of places and nouns of time. When the query data is presented in a table format, OCR software can be used to extract the text and values ​​from the table. The extracted text and values ​​are the query text. Then, the BERT tool is used to segment the query text, identify entity words, verbs and adjectives, and form the first key data group. If the query data is expressed in speech form, speech recognition software is used to extract the text from the speech. The extracted text is the query text. Then, the BERT tool is used to segment the query text, identify entity words, verbs and adjectives, and form the first key data group. If the query data is expressed in the form of an image or video, image processing software is used to describe the text in the image. The described text is the query text. Then, the BERT tool is used to segment the query text, identify entity words, verbs and adjectives, and form the first key data group. In this way, by obtaining the expression form of the query data and extracting keywords based on the expression form of the query data, the efficiency of generating the first key data group can be improved.

[0049] Optionally, responding to the query request based on the second key data group includes: Based on the second key data group, obtain the target data in the target knowledge base that corresponds to the second key data group; Output the target data.

[0050] It should be understood that the target knowledge base can be multiple specified indicator bases. Through the generated second key data set, the target data corresponding to the second key data set is obtained, and the query results, i.e., the target data, are output.

[0051] It should be understood that the query method provided in this application embodiment can be applied to a query system, wherein the query system can be an artificial intelligence-based accuracy search assistance system, such as... Figure 2 As shown, the system includes a user interface module, a natural language processing module, an artificial intelligence engine, a data storage module, and a feedback mechanism module.

[0052] The user interface module is used to receive user-input query requests and display search results, and to record user query request data in real time, forming a query dataset. This real-time recording of user query request data provides basic data support for subsequent analysis and helps improve the accuracy of search results.

[0053] The natural language processing module is used to perform semantic analysis on user query requests, extract keywords, and generate key data sets. This provides precise matching criteria for artificial intelligence engines, improving the relevance and usefulness of search results.

[0054] The artificial intelligence engine is used to output search results, and the artificial intelligence engine is connected via a network. A knowledge base, the artificial intelligence engine based on key data groups Search the knowledge base for relevant matching results and generate corresponding matching data sets. By leveraging artificial intelligence technology, information relevant to user queries can be quickly and accurately retrieved from a knowledge base, and relevant and useful search results can be intelligently matched, thereby effectively improving the user experience.

[0055] The data storage module is used to record users' historical interaction behaviors and form an interaction dataset to provide data support for the system's learning and optimization.

[0056] The feedback mechanism module analyzes the accuracy of each matching result based on the user's historical interaction behavior and generates a corresponding accuracy coefficient. and average accuracy value The feedback mechanism module is set with a fixed range of accurate thresholds. Combined with accuracy coefficient and average accuracy value The system evaluates the reference value of matching results and dynamically adjusts the search strategy based on the evaluation results, enabling the system to continuously optimize and adapt to the query needs and behavioral patterns of different users.

[0057] The expression for querying the dataset is: , to Indicates the user's first time to the [number]th The query data includes text, tables, voice, images, and videos. This indicates the time point at which user query data was obtained.

[0058] The key data group The analysis process is as follows: Based on the query dataset, extract user number 1. The query requires data, where if the user's first query... If the query data is expressed in text format, then BERT is used to segment the query text, identify entity words, verbs, and adjectives, and form key data groups. Entity words include nouns referring to people, places, and times; If the user is If the query data is presented in tabular form, OCR software is used to extract the text and values ​​from the table. The extracted text and values ​​constitute the query text. Then, the BERT tool is used to segment the query text, identify entity words, verbs, and adjectives, and assemble them into key data groups. ; If the user is If the query request is expressed in speech, speech recognition software is used to extract the text from the speech. The extracted text is the query text. Then, the BERT tool is used to segment the query text, identify entity words, verbs, and adjectives, and assemble them into key data sets. ; If the user is If the query data is presented in the form of an image or video, image processing software is used to describe the text in the image. This descriptive text is the query text. The BERT tool is then used to segment the query text, identifying entity words, verbs, and adjectives, and assembling them into key data groups. ; Indicates user number The key data set corresponding to the query request data, and the multimodal data processing capability enable the system to more comprehensively understand the user's query needs. No matter what form the user enters the query request, the system can accurately identify and process it, thereby providing more detailed and comprehensive search results to meet the user's diverse information needs.

[0059] The matching data group The analysis process is as follows: Randomly select the first A knowledge base Using natural language processing technology, the first Key data sets for matching information in a knowledge base and the first The matching results in each knowledge base are marked as ; ; In the formula, Indicates user number The matching data group corresponding to the query requirement data includes One matching result.

[0060] The expression for the interactive dataset is: , , to Indicates the user's first time to the [number]th Interaction behavior data, including the number of clicks on the match result, the duration of the match result's dwell time, the scroll depth of the match result, and the feedback rating of the match result. This indicates the duration of time spent acquiring user interaction behavior data, i.e., the total time a user spends browsing all matching results.

[0061] The accuracy coefficient value The calculation process is as follows: Based on the interactive dataset, extract the first... Interaction behavior data , among which, the The data for the first interaction behavior corresponds to the user's first... Matching data group during the second query , No. Secondary interaction behavior data includes user browsing The system provides feedback information for each matching result, and each matching result corresponds to an accuracy coefficient. ; Based on the interactive dataset, the first In the data of the interaction behavior, the first The number of clicks for each matching result is marked as , , will the The dwell time of each matching result is marked as , will the The scroll depth of each matching result is marked as , will the The feedback score for each matching result is tagged as follows: , will the In the data of this interaction behavior, the total time spent by the user browsing all matching results is marked as ; ; In the formula, Indicates the first The click-through rate of each matching result. This indicates the weight based on the percentage of clicks. Indicates the first The proportion of time each matching result is displayed out of the total time. This indicates the weighting based on the percentage of time spent in the room. Indicates the first The browsing speed of each matching result This indicates the weight given to browsing speed. This indicates the weight given to the feedback rating. , , and All are constants, and 1, Indicates according to , , and Weights, calculated to obtain the first In the data of the interaction behavior, the first The accuracy coefficient of each matching result This greatly improves the accuracy and coverage of searches, optimizes the quality of search results, and enables users to obtain the information they need more quickly.

[0062] The average accuracy value The calculation process is as follows: According to the Secondary interaction behavior data, The accuracy coefficients corresponding to each matching result are denoted as follows: , to Indicates the first to the second The accuracy coefficient corresponding to each matching result; ; In the formula, Indicates the first In the data of this interaction behavior, the average accuracy coefficient of all matching results is... .

[0063] The first The accuracy coefficient of each matching result Less than or equal to the average accuracy value When, it indicates the first The reference value of each matching result is low.

[0064] The first The accuracy coefficient of each matching result Greater than the average accuracy value And included in the accurate threshold When, it indicates the first Each matching result has high reference value.

[0065] The feedback mechanism module ranks the matching results from high to low based on their reference value and assigns an accuracy coefficient. Greater than the average accuracy value The matching results are transmitted to the natural language processing module, which will then convert the accuracy coefficient... Greater than the average accuracy value The matching results are combined with the user's query request for semantic analysis to extract keywords and regenerate key data sets. Iterative updates provide users with a better and more personalized search experience.

[0066] Please see Figure 3 , Figure 3 A schematic diagram of the structure of the query device provided in the embodiments of this application is shown in the figure. The device 300 includes: The first generation module 301 is used to receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group. The second generation module 302 is used to generate a second key data group based on the target matching result and the first key data group. The target matching result is a matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. A response module is used to respond to the query request based on the second key data group.

[0067] Optionally, the historical interaction dataset further includes historical interaction behavior data, and the device 300 further includes: The third generation module 304 is used to generate accuracy coefficient values ​​for the multiple matching results based on the historical interaction behavior data.

[0068] Optionally, the device further includes: The fourth generation module is used to perform weighted summation on the interaction behavior data of the multiple matching results based on the historical interaction behavior data, so as to obtain multiple accuracy coefficient values ​​corresponding to the multiple matching results. The acquisition module is used to obtain the average accuracy coefficient value based on the multiple accuracy coefficient values.

[0069] Optionally, the historical interaction data includes at least one of the following: number of clicks, dwell time, scroll depth, and feedback rating.

[0070] Optionally, the first generation module 301 includes: The first acquisition unit is used to acquire the expression form of the query data; An extraction unit is used to extract keywords from the query data based on the expression form of the query data, wherein the keywords include at least one of entity words, verbs, and adjectives; The generation unit is used to generate the first key data group based on the keywords of the query data.

[0071] Optionally, the response module 303 includes: The second acquisition unit is used to acquire target data in the target knowledge base corresponding to the second key data group based on the second key data group; The output unit is used to output the target data.

[0072] It should be noted that the query device provided in this application embodiment is a device capable of executing the above-described query method. Therefore, all implementation methods in the above-described query method embodiments are applicable to this device and can achieve the same or similar beneficial effects. To avoid repetition, this embodiment will not elaborate further. The query device provided in this application embodiment can also be the above-described query system.

[0073] For details, see Figure 4 As shown in the figure, this application embodiment also provides an electronic device, including a bus 401, a transceiver 402, an antenna 403, a bus interface 404, a processor 405, and a memory 406.

[0074] Processor 405, used for: Receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group; Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group.

[0075] exist Figure 4 In this context, a bus architecture (represented by bus 401) is used. Bus 401 can include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 405 and memory represented by memory 406. Bus 401 can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 404 provides an interface between bus 401 and transceiver 402. Transceiver 402 can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 405 is transmitted over a wireless medium via antenna 403, which further receives data and transmits data to processor 405.

[0076] Processor 405 is responsible for managing bus 401 and general processing, and can also provide various functions, including timing, peripheral interface, voltage regulation, power management, and other control functions. Memory 406 can be used to store data used by processor 405 during operation.

[0077] Alternatively, the processor 405 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or a complex programmable logic device (CPLD).

[0078] Optionally, the historical interaction dataset further includes historical interaction behavior data, and the processor 405 is specifically used for: Based on the historical interaction data, accuracy coefficient values ​​are generated for the multiple matching results.

[0079] Optionally, the processor 405 is further specifically used for: Based on the historical interaction behavior data, the interaction behavior data of the multiple matching results are weighted and summed to obtain multiple accuracy coefficient values ​​corresponding to the multiple matching results; Based on the multiple accuracy coefficient values, the average accuracy coefficient value is obtained.

[0080] Optionally, the historical interaction data includes at least one of the following: number of clicks, dwell time, scroll depth, and feedback rating.

[0081] Optionally, the processor 405 is further specifically used for: Obtain the representation of the queried data; Based on the expression form of the query data, keywords of the query data are extracted, and the keywords include at least one of entity words, verbs and adjectives; Based on the keywords in the query data, the first key data group is generated.

[0082] Optionally, the processor 405 is further configured to: Based on the second key data group, obtain the target data in the target knowledge base that corresponds to the second key data group; Output the target data.

[0083] It should be noted that the electronic device provided in this application embodiment is a device capable of executing the above query method. Therefore, all implementation methods in the above query method embodiments are applicable to this electronic device and can achieve the same or similar beneficial effects. To avoid repetition, this embodiment will not elaborate further.

[0084] This invention also provides an electronic device, including: a processor, a memory, and a program stored in the memory and executable on the processor. When the program is executed by the processor, it implements the various processes of the above-described query method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0085] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the above-described query method embodiments and achieves the same technical effects. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0086] This application also provides a computer program product, including computer instructions. When these computer instructions are executed by a processor, they implement the various processes of the above-described query method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0087] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0088] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0089] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A query method, characterized in that, The method includes: Receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group; Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group.

2. The method according to claim 1, characterized in that, The historical interaction dataset also includes historical interaction behavior data. After receiving the query request, the method further includes: Based on the historical interaction data, the accuracy coefficient values ​​of the multiple matching results are generated.

3. The method according to claim 2, characterized in that, The method further includes: Based on the historical interaction behavior data, the interaction behavior data of the multiple matching results are weighted and summed to obtain multiple accuracy coefficient values ​​corresponding to the multiple matching results; Based on the multiple accuracy coefficient values, the average accuracy coefficient value is obtained.

4. The method according to claim 2, characterized in that, The historical interaction data includes at least one of the following: number of clicks, dwell time, scroll depth, and feedback rating.

5. The method according to claim 1, characterized in that, The step of extracting keywords from the query data corresponding to the query request to generate a first key data group includes: Obtain the representation of the queried data; Based on the expression form of the query data, keywords of the query data are extracted, and the keywords include at least one of entity words, verbs and adjectives; Based on the keywords in the query data, the first key data group is generated.

6. The method according to claim 1, characterized in that, The response to the query request based on the second key data group includes: Based on the second key data group, obtain the target data in the target knowledge base that corresponds to the second key data group; Output the target data.

7. A query device, characterized in that, The device includes: Receive a query request, extract keywords from the query data corresponding to the query request, and generate a first key data group; Based on the target matching result and the first key data group, a second key data group is generated. The target matching result is the matching result whose accuracy coefficient value is greater than the average accuracy coefficient value among multiple matching results determined based on the historical interaction dataset. The historical interaction dataset includes historical query data and multiple matching results corresponding to the historical query data. The query request is responded to based on the second key data group.

8. The apparatus according to claim 7, characterized in that, The historical interaction dataset also includes historical interaction behavior data, and the device further includes: The third generation module is used to generate accuracy coefficient values ​​for the multiple matching results based on the historical interaction behavior data.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the query method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the query method as described in any one of claims 1 to 6.

Images