A problem query method, device, equipment and medium

Abstract

Embodiments of the present specification disclose a problem query method, device, equipment and medium. The scheme can include: obtaining problem information submitted by a user; the problem information is related to historical behavior of the user for a target service; determining a data interface corresponding to the problem information according to a pre-established mapping relationship between interfaces and functions; the data interface is an interface of a business system corresponding to the target service; the mapping relationship is obtained by pre-recognizing using a semantic understanding classification model; querying historical data for the target service based on the historical behavior of the user through the data interface to obtain a query result; and feeding back the query result to a terminal of the user.

Description

Technical Field

[0001] This application relates to the field of computer data processing technology, and in particular to a problem query method, apparatus, device and medium. Background Technology

[0002] As terminal applications become increasingly feature-rich and offer more services, the number of daily active users is also growing rapidly. Consequently, the volume of user inquiries and feedback regarding services is also substantial. To facilitate the processing of this user feedback, almost every system has developed a corresponding information query backend for system service personnel to access user behavior-related information and respond to user feedback. However, this approach is labor-intensive, and the speed of feedback needs improvement. Summary of the Invention

[0003] This specification provides a problem query method, apparatus, device, and medium to address the problem of high labor costs in existing problem query methods.

[0004] To solve the above-mentioned technical problems, the embodiments in this specification are implemented as follows:

[0005] This specification provides an example of a problem query method, which may include:

[0006] Obtain information about issues submitted by users; such information is related to the user's historical actions related to the target service.

[0007] Based on the pre-established mapping relationship between interfaces and functions, the data interface corresponding to the problem information is determined; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model for identification;

[0008] Through the data interface, historical data generated based on the user's historical behavior for the target business can be queried to obtain query results;

[0009] The query results are then sent back to the user's terminal.

[0010] This specification provides an embodiment of a problem query device, which may include:

[0011] The issue acquisition module is used to acquire issue information submitted by users; the issue information is related to the user's historical behavior related to the target business.

[0012] The interface determination module is used to determine the data interface corresponding to the problem information based on a pre-established mapping relationship between interfaces and functions; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model.

[0013] The data query module is used to query historical data for the target business based on the user's historical behavior through the data interface, and obtain query results;

[0014] The result feedback module is used to send the query results back to the user's terminal.

[0015] This specification provides an embodiment of a problem query device, which may include:

[0016] At least one processor; and,

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to:

[0019] Obtain information about issues submitted by users; such information is related to the user's historical actions related to the target service.

[0020] Based on the pre-established mapping relationship between interfaces and functions, the data interface corresponding to the problem information is determined; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model for identification;

[0021] Through the data interface, historical data generated based on the user's historical behavior for the target business can be queried to obtain query results;

[0022] The query results are then sent back to the user's terminal.

[0023] This specification provides an embodiment of a computer-readable medium storing computer-readable instructions that can be executed by a processor to implement a problem query method.

[0024] At least one embodiment of this specification can achieve the following beneficial effects: by establishing a pre-established mapping relationship between interfaces and functions, the data interface corresponding to the obtained user-submitted problem information is determined; the historical data of the target business generated by the user's historical behavior is queried through the data interface; and the query results are fed back to the user's terminal. The existing services of the system can be called by using the pre-identified mapping relationship of the semantic understanding classification model to query the problem and feed it back to the client, thereby automating the problem query, greatly reducing the consumption of manual costs, and improving the speed of problem feedback. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments or prior art of this specification, the drawings used in the description of the embodiments or prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the overall solution architecture of a problem query method provided in the embodiments of this specification in a practical application scenario;

[0027] Figure 2 This is a flowchart illustrating a problem query method provided in an embodiment of this specification;

[0028] Figure 3 This is a swimlane diagram of a problem query method provided in the embodiments of this specification;

[0029] Figure 4 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of the structure of a problem query device;

[0030] Figure 5 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of the structure of a problem query device. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this specification, and not all of them. Based on the embodiments in this specification, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of one or more embodiments of this specification.

[0032] The technical solutions provided in the various embodiments of this specification are described in detail below with reference to the accompanying drawings.

[0033] In existing technologies, the daily active user count for various business functions in terminal applications is relatively large, resulting in a correspondingly high volume of user inquiries and feedback regarding these functions. To facilitate the handling of this user feedback, almost every system has developed a corresponding information query backend for system service personnel to query user behavior-related information in order to respond to user feedback. Since the business systems and the query platform are independent and unrelated, each time a new query scenario is needed, dedicated code for backend querying must be developed to implement the corresponding functionality. However, in most cases, these functionalities are existing system capabilities that simply cannot be simultaneously supported by business calls and backend queries under the current call chain. Furthermore, there are many business systems in the terminal application, and each system develops and maintains its own backend, resulting in high overall development and maintenance costs. The learning curve for system service personnel also varies depending on the backend they operate, placing certain demands on them and thus increasing labor costs.

[0034] To address the shortcomings of existing technologies, this solution provides the following embodiments:

[0035] Figure 1 This is a schematic diagram of the overall architecture of a problem query method provided in the embodiments of this specification in a practical application scenario.

[0036] like Figure 1 As shown, this solution mainly includes: user terminal 1, server 2, and business system server 3. Users can submit problem information on user terminal 1. User terminal 1 can send the submitted problem information to server 2. Server 2 can perform semantic recognition on the problem information, determine the corresponding data interface based on the semantic recognition results, and access business system server 3 through the data interface to obtain the query results. Server 2 can then organize the query results and send them back to user terminal 1 for user viewing.

[0037] Figure 2 This is a flowchart illustrating a problem query method provided in an embodiment of this specification. From a programming perspective, the entity executing the process can be a program or application client hosted on a terminal application server or a mini-program server; specifically, it can be a server used to process user problem information.

[0038] like Figure 2 As shown, the process may include the following steps:

[0039] Step 202: Obtain the problem information submitted by the user; the problem information is related to the user's historical behavior in relation to the target business.

[0040] In the embodiments of this specification, the method of user questioning can be selected from preset questions or entered by the user. Question information can be query-type questions, such as querying the number of times a user has performed a certain operation. Query-type questions can be information related to the user's previous operations, such as the user's historical behavior of clicking a function in a terminal application or mini-program, browsing the interface of a terminal application or mini-program, etc. For example, in some public welfare applications, the terminal application or mini-program may be for virtual feeding of small animals, virtual watering of plants, virtual energy collection, or viewing and processing of user activity—all requiring user operation. The server of the business system providing this service records the user's operation information, or each user operation can be recorded in the system database.

[0041] Step 204: Based on the pre-established mapping relationship between interfaces and functions, determine the data interface corresponding to the problem information; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model.

[0042] The functions in the mapping relationships in the embodiments of this specification can represent the functions that the business system interface itself has in the business system. The mapping relationship can be used to represent the relationship between each data interface API in the business system and its corresponding function. The server can parse the functions of each interface of the business system through a network model with semantic understanding classification capabilities. Based on the parsing results, the server can determine which preset function category each interface belongs to, and then establish a mapping relationship between the interface and its preset function category. It can be understood that the preset function category can be obtained by mining the functions of each existing business in the terminal application or mini-program through a mining model, or it can be obtained by analyzing the code of the written business data interface, or it can be preset by the backend developers based on the functions that each business can implement. For example, in a virtual chicken-feeding mini-program or terminal application, there are interfaces for querying egg donation records and interfaces for querying the number of eggs donated. Assuming that the preset function category includes the function category for querying egg donation status, the semantic understanding classification model can establish a mapping relationship between the above two interfaces and the function category. Assuming that the preset function category includes the function category for querying egg donation records and the function category for querying the number of eggs donated, each interface can also be mapped to its corresponding function category. The mapping relationship described above can be a one-to-one mapping or a many-to-many mapping, depending on the functionality of the actual business system used; no specific limitation is made here. In practical applications, when semantic understanding and recognition fail, backend personnel can be prompted to perform manual verification, and the corresponding target data interface can be obtained based on the result of the manual verification.

[0043] Step 206: Through the data interface, query the historical data for the target business generated based on the user's historical behavior to obtain the query results.

[0044] In the embodiments of this specification, historical data generated by user actions such as operations performed by the user on different business operations provided by the business system can be stored in different databases for easy querying using the corresponding data interface; alternatively, historical data generated by user actions such as operations performed by the user on different business operations provided by the business system can be stored in the same database, and historical data of the target business can be obtained when querying using the corresponding data interface.

[0045] Step 208: Send the query results back to the user's terminal.

[0046] The query results in the embodiments of this specification can be a table of general information; it can also be a table of specific details from the historical data; or it can be a paragraph containing the query data, which is organized by the server according to preset rules, making the information seen by the user more aesthetically pleasing and more in line with language rules. For example: Dear User A, after Xiao A's efforts in querying, we learned that you collected 80 energy on July 5th at 13:20 and 300 energy on July 5th at 20:05. Please check your accounts. We look forward to serving you next time. Goodbye.

[0047] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification may be interchanged according to actual needs, or some steps may be omitted or deleted.

[0048] Figure 2 The method described above determines the data interface corresponding to the user-submitted problem information by establishing a pre-established mapping relationship between interfaces and functions. The data interface queries the historical data of the user's historical behavior that is generated for the target business, and the query results are fed back to the user's terminal. By using the mapping relationship identified in advance by a semantic understanding classification model, the system's existing services can be called to query the problem and feed it back to the client, thereby automating the problem query, greatly reducing the consumption of manual costs, and improving the speed of problem feedback.

[0049] based on Figure 2 In addition to the method described herein, this specification also provides some specific implementation schemes of the method, which will be described below.

[0050] In the embodiments of this specification, computer semantic understanding can be used to determine the purpose or intent of the user's question, and then retrieve results that meet the user's needs. Optionally, the determination of the data interface corresponding to the question information based on a pre-established mapping relationship between interfaces and functions, as described in the embodiments of this specification, may specifically include:

[0051] The question information is semantically understood and identified to determine the user intent corresponding to the question information;

[0052] Based on the pre-established mapping relationship between interfaces and functions, a data interface corresponding to the user intent is determined; the data interface is used to query historical data related to the user intent; the mapping relationship includes a mapping relationship used to represent the correspondence between user intent and data interface.

[0053] In the embodiments of this specification, semantic understanding and recognition can refer to performing Natural Language Processing (NLU) on the question information so that the server can receive the meaning expressed by the question information. User intent can represent the purpose of the user's question; information such as the category of the question the user wants to query and the query object can be determined based on the user intent. The user intent can be obtained by extracting characters related to the user's purpose from the user's question information. The mapping relationship between user intent and data interface can be established through historical question query data.

[0054] To improve the accuracy of the query results, queries can also be performed based on the entity parameter information obtained from entity recognition. Optionally, the querying of historical data related to the target business generated based on the user's historical behavior through the data interface, as described in the embodiments of this specification, to obtain query results, may specifically include:

[0055] Entity recognition is performed on the problem information to determine the entity parameter information contained in the problem information;

[0056] Based on the entity parameter information, the system queries historical data for the target business generated based on the user's historical behavior through the data interface to obtain the query results corresponding to the entity parameter information.

[0057] In the embodiments of this specification, the entity parameter information can represent the entity information contained in the question information, such as time, object, scene, place name, etc. In practical applications, it can be identified according to preset entity categories. The server can also determine the data interface required to query historical data based on certain information in the entity parameters that can represent business functions. The server can determine the data interface corresponding to each entity parameter, and select the interface with the highest matching degree from the corresponding data interfaces as the data interface for querying question information. If the entity information containing each input parameter in the interface is taken as the first entity information, and the entity information in the question information is taken as the second entity information, the matching degree can represent the ratio between the first entity information and the second entity information, and the interface with the highest ratio is taken as the data interface for querying the question.

[0058] To clearly describe the meaning of entity parameters, further limitations have been imposed on them. Optionally, the entity parameter information described in the embodiments of this specification may include entity parameter attributes and parameter attribute values ​​corresponding to the entity parameter attributes; the entity attributes include at least one of business scenario type, start time, and end time.

[0059] In the embodiments of this specification, entity parameter attributes may also include user identifiers and target query quantities, among other entity parameter attributes. Entity parameter attributes can represent entity types; entity parameter attribute values ​​can represent specific data identified from the problem information based on the entity parameter attributes. For example, when the information contains time, time can be used as an entity parameter; when the interface information contains quantity, quantity can be used as entity parameter information.

[0060] In this embodiment, taking entity parameter attributes including scene type, user ID, start time, and end time as an example, if a user requests to query the number of times animals were fed during the period from 15:00 to 17:00 on June 5th, entity recognition can obtain entity parameters such as "scene type = feeding animals", "user ID = 2088000000000001", "starttime = June 5th 15:00", and "end time = June 5th 17:00". Assuming the server can determine the interface to be called as data interface B based on the scene type in the entity parameters, data interface B can be used to query user behavior information generated in the business system during the period from 15:00 to 17:00 on June 5th, and the number of times animals were fed can be counted based on this user behavior information. Data interface B can be understood as the interface in the business system used to count the number of times animals were fed based on user behavior information; it is a data interface that originally existed in the business system. The user ID can be included in the issue information submitted by the user. The user submits the issue information through a terminal application, and the server can also determine the user ID based on the terminal or the terminal application account information used to submit the issue.

[0061] In practical applications, if entity parameter information can include information representing the start and end times of the query, then the query results can include user behavior information generated by the business system within the time period between the start and end times. If there are other restrictive entity parameter attribute values, the server can filter the user behavior information within the time period to obtain entity parameter attribute values ​​that meet other restrictions. Taking virtual egg donation as an example, if a user queries the number of times they donated more than 10 eggs at a time between 9:00 yesterday and 9:00 today, then the server can determine the first user behavior information between the start and end times based on the extracted entity parameter information, thus obtaining user behavior information for those who donated more than 10 eggs at a time within the time period between 9:00 yesterday and 9:00 today.

[0062] To ensure accurate query results based on the mapping relationship, a mapping relationship between the data interface and the functional intent can be pre-established. Optionally, the method described in the embodiments of this specification may further include:

[0063] The code of the business system is scanned to obtain the interface information of each data interface in the business system; the interface information includes at least one of the following: interface name, input parameters, and comments;

[0064] The interface information is identified using a pre-trained first semantic understanding classification model to determine the intent label corresponding to each data interface; the intent label is used to represent the business processing function of the data interface.

[0065] Establish a first mapping relationship between each data interface and the corresponding intent tag.

[0066] In the embodiments described in this specification, the server's code scanning can be a static scan, making the scanning results more accurate. The input parameters for the interface information include time, quantity, and other input parameters. Annotations can be used to annotate and explain some functions or meanings of the interface. The first semantic understanding classification model can be a BiLSTM+CRF model, trained based on existing interface and function data of the business system to better suit business needs. Intent tags can be preset based on interface functions and other information; they can be mined through network models or set through expert experience. Intent tags can be used to represent the function of the data interface.

[0067] This specification allows different data interfaces to be mapped to different intent tags, or related data interfaces to be grouped into a single intent tag, thus facilitating the establishment of a primary mapping relationship between each data interface and that intent tag. It is understood that an interface can be assigned to different intent tags. For example, if interface C is an interface with energy harvesting query functionality, intent tag 1 is for querying energy harvesting flow data, and intent tag 2 is for querying the total energy value collected, then interface C can be assigned to both intent tags 1 and 2, thus establishing a mapping relationship with both and avoiding query failures due to the lack of a mapping relationship. Similarly, assuming interface D is for querying energy harvesting flow data and interface E is for querying the total energy value collected, interface D can be assigned to intent tag 1 and interface E to intent tag 2 during recognition, thus establishing a mapping relationship between interface D and intent tag 1, and between interface E and intent tag 2.

[0068] As one implementation method, optionally, the method of determining the data interface corresponding to the problem information based on a pre-established mapping relationship between interfaces and functions, as described in the embodiments of this specification, may specifically include:

[0069] The question information is identified using a pre-trained second semantic understanding classification model to determine the target intent label corresponding to the question information; the target intent label is used to represent the query business that the question information is targeting.

[0070] Based on the first mapping relationship, the target data interface corresponding to the target intent tag is determined.

[0071] The second semantic understanding classification model in this embodiment can be trained based on historical question information, corresponding historical query results, and other data. The target intent tag can represent the user intent expressed in the question information submitted by the user, i.e., the business the user wants to query. The first mapping relationship can include the correspondence between various data interfaces and intent tags. In this embodiment, the data interface of the business system corresponding to the identified intent tag of the user-submitted question information can be determined to achieve data query.

[0072] In the embodiments of this specification, the second semantic understanding classification model may be the same as or different from the first semantic understanding classification model; no specific limitation is made here. The intent tags in the classification sets corresponding to the second semantic understanding classification model and the first semantic understanding classification model may be the same, and the categories they classify may be consistent to facilitate better data querying. In practical applications, the scenario obtained from entity recognition can also be combined to determine the interface to be accessed. When a business system corresponds to multiple business scenarios, the intent tag may contain information representing the business scenario, thereby enabling the determination of the target data interface corresponding to the target intent tag based on the business scenario.

[0073] As one implementation, optionally, the intent tag described in the embodiments of this specification may include at least one of a first intent tag for representing a statistical quantity query for a first service and a second intent tag for representing an operation flow query for a second service.

[0074] In the embodiments of this specification, intent tags can represent the intent tags of various services provided by the business system. Users can perform certain actions through terminal applications to process the business services provided by the business system. The aforementioned first service and second service can represent the same service or different services. In the embodiments of this specification, intent tags can include tags representing statistical categories or tags representing transaction logs. Specifically, they can also include intent tags used to represent access count queries.

[0075] In practical applications, to improve versatility, the same intent tag can be used to represent business functions in different mini-programs. For example, business function 1 in mini-program A, which counts the number of donated eggs, and business function 2 in mini-program B, which counts the number of energy collected, can both be represented by an intent tag that indicates a query for counting quantities. Of course, to better suit business needs, different business systems can also correspond to different intent tags. No specific limitations are made here.

[0076] To improve the accuracy of query results, the interface information described in the embodiments of this specification includes the input parameters of the data interface; the method may also include:

[0077] Determine the entity labels corresponding to the input parameters in each data interface;

[0078] Establish a second mapping relationship between each input and its corresponding entity label.

[0079] The entity tags in the embodiments of this specification can be pre-set based on the input parameter information in each data interface, and then a second mapping relationship can be established based on the correspondence between the two. For example, the input parameter information in each data interface can be start time, end time, user ID, quantity, transaction record, collection and feeding information, etc. Then, entity tags can be established to correspond one-to-one, so that the corresponding input parameter can be determined according to the entity tag, and then the corresponding data interface can be called to query the corresponding input parameter information.

[0080] To improve the accuracy of query results, this embodiment of the specification may also determine the interface input parameters to be used for data querying based on the entity information in the question information. Optionally, the querying of historical data for the target business based on the user's historical behavior, as described in this embodiment of the specification, through the data interface to obtain query results, may specifically include:

[0081] Entity recognition is performed on the problem information to determine the entity parameter information contained in the problem information;

[0082] Determine the target entity label corresponding to the entity parameter information; the entity label is used to represent the entity attribute of the entity parameter information.

[0083] Based on the second mapping relationship, determine the target input parameters corresponding to the target entity label;

[0084] Based on the target input parameters, the system queries historical data for the target business generated based on the user's historical behavior through the data interface to obtain the query results corresponding to the entity parameter information.

[0085] In the embodiments of this specification, entity recognition models, such as BiLSTM+CRF models, can be used to perform entity recognition on the question information and determine the entity parameter information contained in the question information, such as time information, scene information, etc. In the embodiments of this specification, the entity parameter information may include entity attribute values, such as 1 p.m. In the embodiments of this specification, entity labels can be predefined, and the corresponding entity labels can be determined according to the entity parameter information. The entity labels can represent the entity attributes of the entity attribute values, such as time.

[0086] In the embodiments of this specification, when calling the data interface, the target input parameters can be used to filter historical data of the target business. For example, if the target input parameters are "start time = July 13, 2023, 13:00" and "end time = July 14, 2023, 11:00", the data located between these two time points can be obtained as the query result data. If the target input parameters also contain "Numb = 10", then data with a quantity of 10 located between the two time points can be obtained as the query result.

[0087] To improve the accuracy of recognition, the model can be trained based on real historical data or system data. Optionally, the method described in the embodiments of this specification may further include:

[0088] Obtain interface information of at least some data interfaces in the business system;

[0089] The interface information of at least some of the data interfaces is used as the first training sample to train the first semantic understanding classification model to be trained, thereby obtaining the first semantic understanding classification model for recognizing interface intent.

[0090] And / or,

[0091] Obtain historical problem information submitted by service users of the business system;

[0092] The historical problem information is used as training samples to train the second semantic understanding classification model, resulting in a second semantic understanding classification model for recognizing user intent.

[0093] The data interface information in the embodiments of this specification may include some parameter information of the interface. The semantic understanding classification model can perform natural language processing on the interface information so that the model can understand the physical meaning of the interface information, and then determine the function of the interface based on the interface information, thus using the interface function as the interface intent. For example, whether the interface is used for a query of statistical quantity or a query of user operation transaction history.

[0094] In the embodiments of this specification, when training the second semantic understanding classification model, queries of statistical quantity type and queries of user operation flow type can be classified as two categories to classify historical question information. Then, based on the accuracy or loss of the classification result, it can be determined whether to continue training the model. When the accuracy of the classification result meets the first preset threshold or the loss value meets the second preset threshold, the training of the model can be stopped and the model can be put into use.

[0095] To more clearly illustrate the problem query method provided in the embodiments of this specification, Figure 3 This is a swimlane diagram of a problem query method provided in the embodiments of this specification. For example... Figure 3 As shown, this method can include a business system access phase, a problem acquisition phase, and an information query phase, specifically including:

[0096] Step 302: Scan the code of the business system to obtain the name, input parameters and comments of the business system's interfaces.

[0097] The code of the business system in the embodiments of this specification can represent the code written by the business development user based on the business functions. The code may include interface information of the data interface corresponding to each business.

[0098] In the embodiments described in this specification, the names can be codes such as A, B, and C, or names containing the function of the interface, such as energy collection quantity, energy operation log record, etc., or they can be the identifier of the interface. It is understood that each interface name is unique. Input parameters can be parameter information such as time, quantity, etc. Comments can be explanations or annotations of the interface and input parameters. The names, input parameters, and comments of the interfaces in the business system can all be obtained based on the capabilities of OneAPI.

[0099] Step 304: Identify the interface intent of each interface and establish the first mapping relationship between interface name and interface intent.

[0100] In the embodiments described in this specification, historical business data corresponding to the business data interface can be obtained. The recognition model can identify the interface intent based on the historical business data, thereby establishing a mapping relationship between the interface name and the interface intent, completing the binding between the interface name and the interface intent, so as to intuitively understand the interface intent. The interface intent can be pre-set according to the business that can be processed in the business system.

[0101] Step 306: The server identifies the meaning of each input parameter and establishes a second mapping relationship between the input parameters and the preset entity labels.

[0102] In the embodiments described in this specification, entity labels can be pre-set based on the input parameters of some or all business interfaces. The problem processing server can use a semantic understanding model to perform Natural Language Understanding (NLU) processing on each input parameter to obtain the meaning of each input parameter, thereby establishing a mapping relationship and enabling access from the business system. The server can store a first mapping relationship and a second mapping relationship, as well as the correspondence between each interface and entity parameter information, so that when processing a problem, the corresponding data interface can be called according to the mapping relationship for processing.

[0103] Step 308: The terminal obtains the problem information submitted by the user.

[0104] In practical applications, users can operate on the problem handling page provided by the terminal application and submit problem information.

[0105] The operations described in this specification can be: the user selects one question from several questions displayed on the question submission page and submits it; the user can also edit the question information themselves and submit it; or the user can select from several pieces of information provided on the question submission interface, such as selecting the time period for the query or the type of query.

[0106] Step 310: The terminal sends the user's problem information to the problem handling server.

[0107] Step 312: The server obtains the user-submitted question information sent by the terminal.

[0108] The user-submitted problem information in the embodiments of this specification can be edited by the user. For example, the user reported that they checked in on May 27th and 28th, but did not receive fertilizer. Please verify.

[0109] Step 314: The server identifies the problem information and obtains the user's intent.

[0110] In the embodiments of this specification, the user intent can be an intent tag assigned after the question information is processed by a semantic understanding classification model to obtain the physical meaning of the question information. Continuing the previous example, the user intent could be to query the fertilizer collection record, and the assigned intent tag could be "feed collection record".

[0111] Step 316: The server determines the target data interface based on the user's intent and the first mapping relationship.

[0112] In the embodiments of this specification, the server can use an intent recognition model such as an LSTM+attention model to determine the target data interface that can realize the user's intent based on the mapping relationship between intent tags and data interfaces. Continuing the previous example, the interface determined by the intent tag and the first mapping relationship is interface A1 for querying fertilizer collection flow.

[0113] Step 318: The server uses an entity recognition model to identify the problem information and obtain entity labels.

[0114] The entity recognition model in the embodiments of this specification can perform entity recognition on the question information, obtain different entity parameter information, and obtain entity labels based on the entity parameter information. Continuing the previous example, entity labels such as time, object, and operation can be identified from it.

[0115] Step 320: The server determines the target input parameters based on the entity label and the second mapping relationship.

[0116] In the embodiments of this specification, the target entity parameter information corresponding to the entity label in the entity parameter information of the target data interface can be determined according to the mapping relationship. It can be understood that the target entity parameter information is the target input parameter. Continuing the previous example, the target input parameters obtained according to the mapping relationship are "Time = May 27th and 28th", "Object = Fertilizer" and "Operation = Receipt".

[0117] Step 322: The business system obtains the query results based on the target input parameters and the target data interface.

[0118] In the embodiments described in this specification, the business system server can query historical data of the corresponding target business that meets the target entity parameter information based on the target data interface, organize the data, and obtain the query results. Historical data can be historical behavioral data obtained from user operations on the business. Continuing the previous example, the query results could be: 2023.5.27 13:00 - Check-in +100g, 2023.5.27 13:04 - Browsing Tasks +500g, 2023.5.28 9:00 - Check-in +150g.

[0119] Step 324: The server obtains the query result information sent by the business system.

[0120] In the embodiments described in this manual, the problem-handling server can initiate an RPC call to the business system and directly call back the query results from the business system. The server can organize the query results according to the terminal's preset display conditions to obtain the query result information. Continuing the previous example, if the terminal is set to display AI intelligent dialogue, then the query result information organized by the server is as follows: Dear user, after the efforts of Little Ant, the following results have been obtained: 2023.5.27 13:00 - Check-in +100g, 2023.5.27 13:04 - Browsing Task +500g, 2023.5.28 9:00 - Check-in +150g. These have been distributed, and there are no problems. If you have any further questions, Little Ant is ready to assist you.

[0121] Step 326: The terminal displays the query results sent by the server.

[0122] In the embodiments described in this specification, the terminal can display query results compiled by the server according to preset display conditions. The preset display conditions can be in tabular form, text form, or a combination of both, for user viewing.

[0123] Using the above methods, the system can perform queries based on problem information without building a new query platform. Furthermore, it can achieve automatic queries based on semantic understanding classification models without human intervention, which can reduce labor costs and resource consumption, and also greatly expand the service capabilities of the business system.

[0124] Based on the same idea, embodiments of this specification also provide apparatus corresponding to the above methods. Figure 4 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of the structure of a problem query device. For example... Figure 4 As shown, the device may include:

[0125] Problem acquisition module 402 is used to acquire problem information submitted by the user; the problem information is related to the user's historical behavior in relation to the target business.

[0126] The interface determination module 404 is used to determine the data interface corresponding to the problem information based on the pre-established mapping relationship between interfaces and functions; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model.

[0127] The data query module 406 is used to query historical data for the target business based on the user's historical behavior through the data interface, and obtain query results;

[0128] The result feedback module 408 is used to send the query results back to the user's terminal.

[0129] Based on the same idea, this specification also provides devices corresponding to the above methods in its embodiments.

[0130] Figure 5 The embodiments provided in this specification correspond to Figure 2 A schematic diagram of the structure of a problem query device. For example... Figure 5 As shown, device 500 may include:

[0131] At least one processor 510; and,

[0132] Memory 530 communicatively connected to the at least one processor; wherein,

[0133] The memory 530 stores instructions 520 that can be executed by the at least one processor 510, the instructions being executed by the at least one processor 510 to enable the at least one processor 510 to:

[0134] Obtain information about issues submitted by users; such information is related to the user's historical actions related to the target service.

[0135] Based on the pre-established mapping relationship between interfaces and functions, the data interface corresponding to the problem information is determined; the data interface is the interface of the business system corresponding to the target business; the mapping relationship is obtained in advance by using a semantic understanding classification model for identification;

[0136] Through the data interface, historical data generated based on the user's historical behavior for the target business can be queried to obtain query results;

[0137] The query results are then sent back to the user's terminal.

[0138] Following the same approach, embodiments of this specification also provide a computer-readable medium corresponding to the above-described method. The computer-readable medium stores computer-readable instructions, which can be executed by a processor to implement the above-described problem query method:

[0139] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, for... Figure 5 As the device shown is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0140] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program and "integrate" a digital system onto a PLD themselves, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software. Similar to the software compiler used in program development, the original code before compilation must be written in a specific programming language, called a Hardware Description Language (HDL). There are many HDLs, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should understand that by simply performing some logic programming on the method flow using one of these hardware description languages ​​and programming it into an integrated circuit, the hardware circuit implementing the logical method flow can be easily obtained.

[0141] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0142] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0143] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0144] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0145] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0146] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0147] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0148] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0149] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0150] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0151] It should also be noted that 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. Without further limitation, 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 said element.

[0152] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0153] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0154] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A problem query method, applied to a server, including: The code of the business system is scanned to obtain the interface information of each data interface in the business system. The interface information includes at least one of the following: interface name, input parameters, and comments. The interface information is identified using a pre-trained first semantic understanding classification model to determine the intent tag corresponding to each data interface. The intent tag is used to represent the business processing function of the data interface. A first mapping relationship between each data interface and the corresponding intent tag is established. The entity tag corresponding to the input parameters in each data interface is determined. A second mapping relationship between each input parameter and the corresponding entity tag is established. Obtain information about issues submitted by users; such information is related to the user's historical actions related to the target service. Based on a pre-established mapping relationship between interfaces and functions, the data interface corresponding to the question information is determined. This determination includes: using a pre-trained second semantic understanding classification model to identify the question information and determine the target intent tag corresponding to the question information. The target intent tag represents the query business that the question information targets. Based on the first mapping relationship, the target data interface corresponding to the target intent tag is determined. The data interface is the interface of the business system corresponding to the target business, used to access the business system server. The mapping relationship is obtained in advance using a semantic understanding classification model, and the first mapping relationship includes the correspondence between each data interface and the intent tag. Through the data interface, the business system server queries historical data generated based on the user's historical behavior for the target business to obtain query results. This process includes: performing entity recognition on the problem information; determining the entity parameter information contained in the problem information; determining the target entity tag corresponding to the entity parameter information; the entity tag representing the entity attribute of the entity parameter information; determining the target input parameter corresponding to the target entity tag according to the second mapping relationship; and, based on the target input parameter, querying historical data generated based on the user's historical behavior for the target business through the data interface to obtain query results corresponding to the entity parameter information. The query results are then sent back to the user's terminal.

2. The method according to claim 1, wherein determining the data interface corresponding to the problem information based on a pre-established mapping relationship between interfaces and functions specifically includes: The question information is semantically understood and identified to determine the user intent corresponding to the question information; Based on the pre-established mapping relationship between interfaces and functions, a data interface corresponding to the user intent is determined; the data interface is used to query historical data related to the user intent; the mapping relationship includes a mapping relationship used to represent the correspondence between user intent and data interface.

3. The method according to claim 1, wherein querying historical data for the target business based on the user's historical behavior on the business system server through the data interface to obtain query results specifically includes: Entity recognition is performed on the problem information to determine the entity parameter information contained in the problem information; Based on the entity parameter information, the system queries historical data for the target business generated based on the user's historical behavior through the data interface to obtain the query results corresponding to the entity parameter information.

4. The method according to claim 3, wherein the entity parameter information includes entity parameter attributes and parameter attribute values ​​corresponding to the entity parameter attributes; the entity attributes include at least one of business scenario type, start time, and end time.

5. The method according to claim 1, wherein the intent tag includes at least one of a first intent tag for representing a statistical quantity query for a first service and a second intent tag for representing an operation flow query for a second service.

6. The method according to claim 1, further comprising: Obtain interface information of at least some data interfaces in the business system; The interface information of at least some of the data interfaces is used as the first training sample to train the first semantic understanding classification model to be trained, thereby obtaining the first semantic understanding classification model for recognizing interface intent. And / or, Obtain historical problem information submitted by service users of the business system; The historical problem information is used as training samples to train the second semantic understanding classification model, resulting in a second semantic understanding classification model for recognizing user intent.

7. A problem query device, applied to a server, comprising: The issue acquisition module scans the code of the business system to obtain interface information for each data interface in the business system. This interface information includes at least one of the following: interface name, input parameters, and comments. A pre-trained first semantic understanding classification model is used to identify the interface information, determining the intent tag corresponding to each data interface. The intent tag represents the business processing function of the data interface. A first mapping relationship is established between each data interface and its corresponding intent tag. The module also determines the entity tags corresponding to the input parameters in each data interface and establishes a second mapping relationship between each input parameter and its corresponding entity tag. Furthermore, the module acquires user-submitted issue information, which is related to the user's historical actions related to the target business. An interface determination module is used to determine the data interface corresponding to the question information based on a pre-established mapping relationship between interfaces and functions. This determination includes: using a pre-trained second semantic understanding classification model to identify the question information and determine a target intent tag corresponding to the question information. The target intent tag represents the query business that the question information targets. Based on the first mapping relationship, a target data interface corresponding to the target intent tag is determined. The data interface is the interface of the business system corresponding to the target business, used to access the business system server. The mapping relationship is obtained in advance using a semantic understanding classification model, and the first mapping relationship includes the correspondence between each data interface and the intent tag. The data query module is used to query historical data generated based on the user's historical behavior for the target business through the data interface on the business system server to obtain query results. The process of querying historical data generated based on the user's historical behavior for the target business through the data interface to obtain query results includes: performing entity recognition on the problem information; determining the entity parameter information contained in the problem information; determining the target entity tag corresponding to the entity parameter information; the entity tag representing the entity attribute of the entity parameter information; determining the target input parameter corresponding to the target entity tag according to the second mapping relationship; and, based on the target input parameter, querying historical data generated based on the user's historical behavior for the target business through the data interface to obtain query results corresponding to the entity parameter information. The result feedback module is used to send the query results back to the user's terminal.

8. A problem query device, comprising: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, which, when executed by the at least one processor, enable the at least one processor to: The code of the business system is scanned to obtain the interface information of each data interface in the business system. The interface information includes at least one of the following: interface name, input parameters, and comments. The interface information is identified using a pre-trained first semantic understanding classification model to determine the intent tag corresponding to each data interface. The intent tag is used to represent the business processing function of the data interface. A first mapping relationship between each data interface and the corresponding intent tag is established. The entity tag corresponding to the input parameters in each data interface is determined. A second mapping relationship between each input parameter and the corresponding entity tag is established. Obtain information about issues submitted by users; such information is related to the user's historical actions related to the target service. Based on a pre-established mapping relationship between interfaces and functions, the data interface corresponding to the question information is determined. This determination includes: using a pre-trained second semantic understanding classification model to identify the question information and determine the target intent tag corresponding to the question information. The target intent tag represents the query business that the question information targets. Based on the first mapping relationship, the target data interface corresponding to the target intent tag is determined. The data interface is the interface of the business system corresponding to the target business, used to access the business system server. The mapping relationship is obtained in advance using a semantic understanding classification model, and the first mapping relationship includes the correspondence between each data interface and the intent tag. Through the data interface, the business system server queries historical data generated based on the user's historical behavior for the target business to obtain query results. This process includes: performing entity recognition on the problem information; determining the entity parameter information contained in the problem information; determining the target entity tag corresponding to the entity parameter information; the entity tag representing the entity attribute of the entity parameter information; determining the target input parameter corresponding to the target entity tag according to the second mapping relationship; and, based on the target input parameter, querying historical data generated based on the user's historical behavior for the target business through the data interface to obtain query results corresponding to the entity parameter information. The query results are then sent back to the user's terminal.

9. A computer-readable medium having stored thereon computer-readable instructions that can be executed by a processor to implement the problem query method of any one of claims 1 to 6.

Images