Data query method and device, equipment and medium

Abstract

The invention discloses a data query method and device, equipment and a medium. The method comprises the steps that after a question text input by a user is obtained, a question scene of the question text is detected; when detecting that the problem scene is a data query scene, generating a query description text of each piece of target information through a query task splitting model; generating a data query statement of each piece of target information through the data query model, obtaining business data related to each piece of target information from a preset database, and displaying the business data; and generating a data query summary text of the question text through the query summary model, and displaying the data query summary text. According to the embodiment of the invention, the business data related to each piece of target information can be automatically obtained from the business data table in the database on the basis of the large-scale language model and the problem text input when the user is in a scene needing to query the business data, and the business data is provided for the user through the interactive page.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a data query method, device, equipment and medium. Background Art

[0002] The business system of an enterprise usually obtains a large amount of business data during operation. In order to effectively manage the business data, the business system will store the obtained business data in the corresponding business data table in the database set in the data storage device, and record the obtained business data. Business personnel can query the business data they need to obtain in each business data table in the database set in the data storage device.

[0003] In the related art, the commonly used data query scheme is: when it is necessary to obtain business data related to multiple target information, the business personnel respectively query the business data related to each target information in the business data table of the database set in the data storage device, and obtain the business data related to each target information. The target information can be the attribute information of a designated business personnel or a designated product in the enterprise. The business data related to the target information can refer to each business data containing the target information. The data query scheme in the related art needs to be manually operated to query the business data related to each target information in the business data table of the database set in the data storage device. The query efficiency is low, the labor cost and time cost are high, and the accuracy is difficult to guarantee. Summary of the invention

[0004] The present invention provides a data query method, device, equipment and medium to solve the problems of low query efficiency, high labor cost and time cost, and difficulty in ensuring accuracy of data query solutions in related technologies.

[0005] According to one aspect of the present invention, there is provided a data query method, comprising:

[0006] After obtaining the question text input by the user, the question scenario of the question text is detected through the question analysis model;

[0007] When it is detected that the problem scenario of the question text is a data query scenario, a query description text of each target information is generated according to each target information in the question text and data table range information of each target information through a query task splitting model;

[0008] Generate data query statements for each target information based on the query description text of each target information through the data query model, obtain business data related to each target information from a preset database based on the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page;

[0009] Through the query summary model, according to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, and the data query summary text of the question text is displayed in the user interaction page.

[0010] According to another aspect of the present invention, there is provided a data query device, comprising:

[0011] A scene detection module is used to detect the problem scene of the question text through the question analysis model after obtaining the question text input by the user;

[0012] A query task splitting module, for generating a query description text for each target information according to each target information in the question text and the data table range information of each target information through a query task splitting model when it is detected that the problem scenario of the question text is a data query scenario;

[0013] The database query module is used to generate a data query statement for each target information according to the query description text of each target information through the data query model, and obtain the business data related to each target information from the preset database according to the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page;

[0014] The query summary module is used to generate a data query summary text of the question text based on the query description text of each target information and the business data related to each target information through a query summary model, and display the data query summary text of the question text in the user interaction page.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] at least one processor;

[0017] and a memory communicatively coupled to the at least one processor;

[0018] The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data query method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data query method described in any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention is to detect the problem scenario of the problem text through the problem analysis model after obtaining the problem text input by the user; when it is detected that the problem scenario of the problem text is a data query scenario, the query description text of each target information and the data table range information of each target information are generated through the query task splitting model; then, the data query statement of each target information is generated according to the query description text of each target information through the data query model, and the business data related to each target information is obtained from the preset database according to the data query statement of each target information, and the query description text, data query statement and business data related to each target information of each target information are displayed on the user interaction page; finally, the query summary model is used to According to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, and the data query summary text of the question text is displayed in the user interaction page, which solves the problems of low query efficiency, high labor cost and time cost, and difficult to ensure accuracy of data query solutions in related technologies. Based on a large-scale language model and the question text entered by the user when it is necessary to query business data related to multiple target information in the business data table in the database, the business data related to each target information can be automatically queried in the business data table in the database, the business data related to each target information is obtained, and the business data related to each target information is provided to the user through the interactive page, the labor cost and time cost are low, and the accuracy and efficiency of the data query process are improved.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 The present invention provides a flowchart of a data query method according to the first embodiment of the present invention.

[0024] Figure 2 The present invention provides a flowchart of a data query method according to the second embodiment of the present invention.

[0025] Figure 3 A schematic diagram of the structure of a data query device provided in Embodiment 3 of the present invention.

[0026] Figure 4 A schematic diagram of the structure of an electronic device for implementing the data query method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "target", "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprise", "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] Embodiment 1

[0030] Figure 1A flowchart of a data query method provided in Embodiment 1 of the present invention. This embodiment is applicable to the case where business data to be obtained is queried in various business data tables in a database set in a data storage device. The method can be executed by a data query device, which can be implemented in the form of hardware and / or software, and the data query device can be configured in a data storage device. A data storage device is an electronic device set in an enterprise for storing business data. A preset database is set in the data storage device. Figure 1 As shown, the method includes:

[0031] Step 101: After obtaining the question text input by the user, the question scenario of the question text is detected through the question analysis model.

[0032] Optionally, the business system of an enterprise may be a server for managing the business of the enterprise. The business system usually acquires a large amount of business data during operation. The business personnel of the enterprise may be technical personnel responsible for managing business data or products. The products of the enterprise may be software products or hardware products produced by the enterprise. Each business data acquired may be a data related to a specified business personnel, or may be a data related to a specified product. The preset database is a database for storing business data set in a data storage device. A plurality of business data tables are set in the preset database. Each business data table may be a data table for storing business data of a specified type or business data acquired on a specified date. Each business data table is provided with identification information. The identification information of the business data table may be a number or a character string pre-set to uniquely identify the business data table. The business system may store the business data in a corresponding business data table in a database set in the data storage device according to the type of each business data acquired or the acquisition date, and record the business data. The user may be any business personnel.

[0033] Optionally, when the user is in a scenario where the business data table in the preset database needs to query the business data related to multiple target information, the question text entered by the user may be a text for describing whether the user asks the data storage device whether the business data related to multiple target information can be queried in the business data table in the preset database, and the business data related to each target information is provided to the user. The question text entered by the user contains multiple target information and data table range information of each target information. The number of target information contained in the question text entered by the user is greater than or equal to 2. Each target information may be the attribute information of a designated business person or a designated product in the enterprise. The attribute information of the business person is provided by the business person to the business system and other business persons in the enterprise and authorized for use by the business system and other business persons in the enterprise. The attribute information of the business person may be the contact information, identification information or social identity information of the business person. The contact information of the business person may refer to the mobile phone number of the business person. The identification information of the business person may refer to the ID card number of the business person. The social identity information of the business person may refer to the account used by the business person in the social software. The attribute information of the product is provided by the business person responsible for managing the product to the business system and other business persons in the enterprise. The attribute information of the product may be the identification information of the product. The identification information of the product may be a number or a character string that is preset to uniquely identify the product. The business data related to the target information may refer to each business data containing the target information. The data table range information of the target information may be the identification information of the business data table containing the business data related to the target information.

[0034] Optionally, when the user is in a scenario where intent analysis, sentiment analysis, or negative sentiment analysis is required, the question text input by the user may be text used to describe the question the user asked related to intent analysis, sentiment analysis, or negative sentiment analysis.

[0035] Optionally, when the user is in a scenario where a general chat is required, the question text input by the user may be text used to describe a question related to the general chat asked by the user.

[0036] Optionally, the user interaction page may be a page set in the data storage device for interacting with the user, obtaining the question text input by the user, and providing the user with business data queried according to the question text or the answer text of the question text. The user may enter the question text by filling in the question text in the user interaction page. The data storage device may obtain the question text filled in by the user in the user interaction page, thereby obtaining the question text input by the user, and then detect the problem scenario of the question text through the question analysis model.

[0037] Optionally, the problem scenario of the question text input by the user can be a text used to characterize the scenario in which the user enters the question text. The problem scenario of the question text input by the user is a data query scenario, a personnel intention analysis scenario, or a normal chat scenario. If the problem scenario of the question text input by the user is a data query scenario, it indicates that the scenario in which the user enters the question text is a scenario in which business data related to multiple target information needs to be queried in a business data table in a preset database. If the problem scenario of the question text input by the user is a personnel intention analysis scenario, it indicates that the scenario in which the user enters the question text is a scenario in which intent analysis, sentiment analysis, or negative sentiment analysis is required. If the problem scenario of the question text input by the user is a normal chat scenario, it indicates that the scenario in which the user enters the question text is a scenario in which normal chat is required.

[0038] Optionally, a pre-trained problem analysis model is provided in the data storage device. The pre-trained problem analysis model may be a large-scale language model (LLM) pre-trained by a technician and provided in the data storage device for detecting the problem scenario of the problem text input by the user according to the problem analysis prompt instruction. The problem analysis prompt instruction is used to instruct the pre-trained problem analysis model to detect and analyze the problem text input by the user, and output the text of the problem scenario of the problem text input by the user. The problem analysis prompt instruction contains the problem text for which the problem scenario needs to be detected. The problem analysis prompt instruction is input into the pre-trained problem analysis model, and the pre-trained problem analysis model will detect and analyze the problem text for which the problem scenario needs to be detected contained in the problem analysis prompt instruction, and output the problem scenario of the problem text.

[0039] Optionally, the problem scenario of the question text is detected through the question analysis model, including: generating a question analysis prompt instruction corresponding to the question text according to the question text and a question analysis prompt instruction template; inputting the question analysis prompt instruction into a pre-trained question analysis model to obtain the problem scenario of the question text output by the question analysis model; wherein the problem scenario of the question text is a data query scenario, a personnel intention analysis scenario or a normal chat scenario.

[0040] Optionally, the problem analysis prompt instruction template is a problem analysis prompt instruction that does not include the problem text that needs to detect the problem scenario. The problem analysis prompt instruction template includes a text filling position. The text filling position is a position used to fill the problem text that needs to detect the problem scenario. After filling the problem text that needs to detect the problem scenario into the text filling position in the problem analysis prompt instruction template, a problem analysis prompt instruction corresponding to the problem text that needs to detect the problem scenario can be obtained. The problem analysis prompt instruction corresponding to the problem text that needs to detect the problem scenario is used to instruct a pre-trained problem analysis model to detect and analyze the problem text, and output the text of the problem scenario of the problem text. The problem analysis prompt instruction template is stored in the data storage device.

[0041] Optionally, generating a question analysis prompt instruction corresponding to the question text according to the question text and the question analysis prompt instruction template includes: filling the question text into the text filling position in the question analysis prompt instruction template, thereby obtaining a question analysis prompt instruction corresponding to the question text. The question analysis prompt instruction corresponding to the question text is used to instruct a pre-trained question analysis model to detect and analyze the question text and output the text of the question scenario of the question text.

[0042] Optionally, the data storage device inputs the question analysis prompt instruction corresponding to the question text into a pre-trained question analysis model. The pre-trained question analysis model detects and analyzes the question text contained in the question analysis prompt instruction, and outputs the question scenario of the question text. The data storage device obtains the question scenario of the question text output by the question analysis model.

[0043] Step 102: When it is detected that the problem scenario of the question text is a data query scenario, a query description text for each target information is generated according to each target information and data table range information of each target information in the question text through a query task splitting model.

[0044] Optionally, the query description text of the target information may be a text for describing an operation of acquiring the business data related to the target information from a business data table containing the business data related to the target information.

[0045] Optionally, a pre-trained query task splitting model is provided in the data storage device. The pre-trained query task splitting model may be a large-scale language model pre-trained by a technician and provided in the data storage device for generating query description texts for each target information in the question text according to the query task splitting prompt instruction. The query task splitting prompt instruction is a text for instructing the pre-trained query task splitting model to detect and analyze the question text containing each target information and the data table range information of each target information, and generate query description texts for each target information in the question text. The query task splitting prompt instruction contains the question text for which query description texts for each target information need to be generated. The query task splitting prompt instruction is input into the pre-trained query task splitting model, and the pre-trained query task splitting model will detect and analyze the question text for which query description texts for each target information need to be generated contained in the query task splitting prompt instruction, generate query description texts for each target information in the question text, and output the query description texts for each target information in the question text.

[0046] Optionally, a query description text for each target information is generated according to each target information in the question text and data table range information of each target information through a query task splitting model, including: generating a query task splitting prompt instruction corresponding to the question text according to the question text and a query task splitting prompt instruction template; inputting the query task splitting prompt instruction into a pre-trained query task splitting model to obtain a query description text for each target information generated according to each target information in the question text and data table range information of each target information output by the query task splitting model.

[0047] Optionally, the query task splitting prompt instruction template is a query task splitting prompt instruction that does not contain the question text that needs to generate the query description text of each target information. The query task splitting prompt instruction template contains a text filling position. The text filling position is the position of the question text that is used to fill the query description text that needs to generate each target information. After filling the question text that needs to generate the query description text of each target information into the text filling position in the query task splitting prompt instruction template, the query task splitting prompt instruction corresponding to the question text that needs to generate the query description text of each target information can be obtained. The query task splitting prompt instruction corresponding to the question text that needs to generate the query description text of each target information is a text used to instruct the pre-trained query task splitting model to detect and analyze the question text and generate the query description text of each target information in the question text. The query task splitting prompt instruction template is stored in the data storage device.

[0048] Optionally, generating a query task splitting prompt instruction corresponding to the question text according to the question text and the query task splitting prompt instruction template includes: filling the question text into the text filling position in the query task splitting prompt instruction template, thereby obtaining the query task splitting prompt instruction corresponding to the question text. The query task splitting prompt instruction corresponding to the question text is a text used to instruct a pre-trained question analysis model to detect and analyze the question text and generate a query description text for each target information in the question text.

[0049] Optionally, the data storage device inputs the query task splitting prompt instruction corresponding to the question text into a pre-trained query task splitting model. The pre-trained query task splitting model detects and analyzes the question text contained in the query task splitting prompt instruction, generates query description texts for each target information in the question text, and outputs the query description texts for each target information in the question text.

[0050] Step 103: Generate data query statements for each target information based on the query description text of each target information through the data query model, and obtain business data related to each target information from a preset database based on the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page.

[0051] Optionally, the data query statement may be a code statement used to obtain specified business data from a preset database in the data storage device after being executed. The data storage device may obtain the specified business data from the preset database by executing the data query statement. The data query statement for the target information may be a code statement used to obtain business data related to the target information from a business data table containing business data related to the target information in a preset database in the data storage device after being executed. The data storage device may obtain business data related to the target information from the preset database by executing the data query statement for the target information.

[0052] Optionally, a pre-trained data query model is provided in the data storage device. The pre-trained data query model may be a large-scale language model that is pre-trained by a technician and provided in the data storage device for generating a data query statement for the target information according to a query statement generation prompt instruction. The query statement generation prompt instruction is a text for instructing the pre-trained data query model to detect and analyze the query description text of the target information and generate a data query statement for the target information. The query statement generation prompt instruction contains the query description text of the target information for which a data query statement needs to be generated. The query statement generation prompt instruction is input into the pre-trained data query model, and the pre-trained data query model detects and analyzes the query description text of the target information for which a data query statement needs to be generated contained in the query statement generation prompt instruction, generates a data query statement for the target information, and outputs the data query statement for the target information.

[0053] Optionally, a data query statement for each target information is generated through a data query model according to the query description text of each target information, including: generating a prompt instruction template according to the query description text and query statement of each target information, and generating a query statement generation prompt instruction corresponding to each target information; inputting the query statement generation prompt instruction corresponding to each target information into a pre-trained data query model, and obtaining the data query statement for each target information output by the data query model.

[0054] Optionally, the query statement generation prompt instruction template is a query statement generation prompt instruction that does not contain the query description text of the target information for which the data query statement needs to be generated. The query statement generation prompt instruction template contains a text filling position. The text filling position is a position for filling the query description text of the target information for which the data query statement needs to be generated. After filling the query description text of the target information for which the data query statement needs to be generated into the text filling position in the query statement generation prompt instruction template, a query statement generation prompt instruction corresponding to the target information for which the data query statement needs to be generated can be obtained. The query statement generation prompt instruction corresponding to the target information for which the data query statement needs to be generated is a text for instructing a pre-trained data query model to detect and analyze the query description text of the target information to generate the data query statement for the target information. The query statement generation prompt instruction template is stored in the data storage device.

[0055] Optionally, generating a query statement generation prompt instruction corresponding to the target information based on the query description text of the target information and the query statement generation prompt instruction template includes: filling the query description text of the target information into the text filling position in the query statement generation prompt instruction template, thereby obtaining a query statement generation prompt instruction corresponding to the target information. For each target information in the question text, fill the query description text of the target information into the text filling position in the query statement generation prompt instruction template, thereby obtaining a query statement generation prompt instruction corresponding to the target information. The query statement generation prompt instruction corresponding to the target information is used to instruct a pre-trained data query model to detect and analyze the query description text of the target information and generate the text of the data query statement of the target information.

[0056] Optionally, for each target information in the question text, the data storage device inputs the query statement generation prompt instruction corresponding to the target information into the pre-trained data query model. The pre-trained data query model detects and analyzes the query description text of the target information contained in the query statement generation prompt instruction, generates a data query statement for the target information, and outputs the data query statement for the target information.

[0057] Optionally, based on the data query statements of each target information, business data related to each target information is obtained from a preset database, and the query description text, data query statements and business data related to each target information are displayed in the user interaction page, including: performing the following operations for each target information in the question text: running the data query statement of the target information, thereby obtaining the business data related to the target information from the preset database; displaying the query description text, data query statement and business data related to the target information in the user interaction page.

[0058] Step 104: Generate a data query summary text of the question text based on the query description text of each target information and the business data related to each target information through the query summary model, and display the data query summary text of the question text in the user interaction page.

[0059] Optionally, the data query summary text of the question text may be a query description text for describing each target information in the question text and a text of business data related to each target information in the question text. The data query summary text of the question text includes the query description text for each target information in the question text, the total amount of business data included in the business data related to each target information obtained, and the business data related to each target information obtained.

[0060] Optionally, a pre-trained query summary model is provided in the data storage device. The pre-trained query summary model can be a large-scale language model pre-trained by a technician and provided in the data storage device for generating a data query summary text of the question text according to the summary text generation prompt instruction. The summary text generation prompt instruction is used to instruct the pre-trained query summary model to detect and analyze the query description text of each target information in the question text and the business data related to each target information in the question text, and generate a data query summary text of the question text. The summary text generation prompt instruction contains the query description text of each target information in the question text for which the data query summary text needs to be generated, and the business data related to each target information in the question text for which the data query summary text needs to be generated. The summary text generation prompt instruction is input into the pre-trained query summary model, and the pre-trained query summary model will detect and analyze the query description text of each target information in the question text for which the data query summary text needs to be generated, and the business data related to each target information in the question text for which the data query summary text needs to be generated, contained in the summary text generation prompt instruction, to generate a data query summary text of the question text, and output the data query summary text of the question text.

[0061] Optionally, a data query summary text of the question text is generated through a query summary model according to query description texts of each target information and business data related to each target information, including: generating a summary text generation prompt instruction corresponding to the question text according to the query description texts of each target information, business data related to each target information and a summary text generation prompt instruction template; inputting the summary text generation prompt instruction into a pre-trained query summary model to obtain the data query summary text of the question text generated by the query summary model according to the query description texts of each target information and business data related to each target information.

[0062] Optionally, the summary text generation prompt instruction template is a summary text generation prompt instruction that does not contain the query description text of each target information in the question text that needs to generate the data query summary text, and the business data related to each target information in the question text that needs to generate the data query summary text. The summary text generation prompt instruction template contains a text filling position. The text filling position is a position for filling the query description text of each target information in the question text that needs to generate the data query summary text, and the business data related to each target information in the question text that needs to generate the data query summary text. After filling the query description text of each target information in the question text that needs to generate the data query summary text, and the business data related to each target information in the question text that needs to generate the data query summary text into the text filling position in the summary text generation prompt instruction template, the summary text generation prompt instruction corresponding to the question text that needs to generate the data query summary text can be obtained. The summary text generation prompt instruction corresponding to the question text that needs to generate the data query summary text is used to instruct the pre-trained query summary model to detect and analyze the query description text of each target information in the question text and the business data related to each target information in the question text, and generate the data query summary text of the question text. The data storage device stores the summary text generation prompt instruction template.

[0063] Optionally, generating a summary text generation prompt instruction corresponding to the question text according to the query description text of each target information, the business data related to each target information, and the summary text generation prompt instruction template includes: filling the query description text of each target information in the question text and the business data related to each target information in the question text into the text filling position in the summary text generation prompt instruction template, thereby obtaining the summary text generation prompt instruction corresponding to the question text. The summary text generation prompt instruction corresponding to the question text is used to instruct the pre-trained query summary model to detect and analyze the query description text of each target information in the question text and the business data related to each target information in the question text, and generate the data query summary text of the question text.

[0064] Optionally, the data storage device inputs the summary text generation prompt instruction corresponding to the question text into a pre-trained query summary model. The pre-trained query summary model detects and analyzes the query description text of each target information in the question text contained in the summary text generation prompt instruction and the business data related to each target information in the question text, generates a data query summary text of the question text, and outputs the data query summary text of the question text. The data storage device obtains the data query summary text of the question text output by the query summary model, and displays the data query summary text of the question text in the user interaction page.

[0065] Optionally, after detecting the problem scenario of the question text through the question analysis model, it also includes: when it is detected that the problem scenario of the question text is a personnel intention analysis scenario, generating an answer text for the question text through the intention analysis model, and displaying the answer text of the question text in the user interaction page.

[0066] Optionally, if the question scenario of the question text is a personnel intent analysis scenario, the question text is a text used to describe the questions asked by the user related to intent analysis, sentiment analysis or negative sentiment analysis, and the answer text of the question text is a text used to describe the answers to the questions asked by the user described in the question text related to intent analysis, sentiment analysis or negative sentiment analysis.

[0067] Optionally, a pre-trained intent analysis model is provided in the data storage device. The pre-trained intent analysis model can be a large-scale language model pre-trained by a technician and provided in the data storage device for generating an answer text for a question text describing a question related to intent analysis, sentiment analysis, or negative sentiment analysis inquired by a user according to a first text generation prompt instruction. The first text generation prompt instruction is a text for instructing the pre-trained intent analysis model to detect and analyze the question text for describing the question related to intent analysis, sentiment analysis, or negative sentiment analysis inquired by the user, and generate an answer text for the question text. The first text generation prompt instruction contains the question text for describing the question related to intent analysis, sentiment analysis, or negative sentiment analysis inquired by the user, for which an answer text needs to be generated. The first text generation prompt instruction is input into the pre-trained intent analysis model, and the pre-trained intent analysis model will detect and analyze the question text for describing the question related to intent analysis, sentiment analysis, or negative sentiment analysis inquired by the user contained in the first text generation prompt instruction, generate an answer text for the question text, and output the answer text for the question text.

[0068] Optionally, an answer text to the question text is generated through an intention analysis model, including: generating a first text generation prompt instruction corresponding to the question text based on the question text and a first text generation prompt instruction template; inputting the first text generation prompt instruction into a pre-trained intention analysis model to obtain the answer text to the question text output by the intention analysis model.

[0069] Optionally, the first text generation prompt instruction template is a first text generation prompt instruction that does not contain a question text for describing a question related to intent analysis, sentiment analysis, or negative emotion analysis asked by a user, for which an answer text needs to be generated. The first text generation prompt instruction template contains a text filling position. The text filling position is a position for filling the question text for describing a question related to intent analysis, sentiment analysis, or negative emotion analysis asked by a user, for which an answer text needs to be generated. After filling the question text for describing a question related to intent analysis, sentiment analysis, or negative emotion analysis asked by a user, for which an answer text needs to be generated, into the text filling position in the first text generation prompt instruction template, a first text generation prompt instruction corresponding to the question text for describing a question related to intent analysis, sentiment analysis, or negative emotion analysis asked by a user, for which an answer text needs to be generated, can be obtained. The first text generation prompt instruction corresponding to the question text for describing a question related to intent analysis, sentiment analysis, or negative emotion analysis asked by a user, for which an answer text needs to be generated, is a text for instructing a pre-trained intent analysis model to detect and analyze the question text and generate an answer text for the question text. The first text generation prompt instruction template is stored in the data storage device.

[0070] Optionally, generating a first text generation prompt instruction corresponding to the question text according to the question text and the first text generation prompt instruction template includes: filling the question text into the text filling position in the first text generation prompt instruction template, thereby obtaining a first text generation prompt instruction corresponding to the question text. The first text generation prompt instruction corresponding to the question text is a text for instructing a pre-trained intent analysis model to detect and analyze the question text used to describe a question asked by the user related to intent analysis, sentiment analysis, or negative sentiment analysis, and generate an answer text for the question text.

[0071] Optionally, the data storage device inputs the first text generation prompt instruction corresponding to the question text into a pre-trained intent analysis model. The pre-trained intent analysis model detects and analyzes the question text contained in the first text generation prompt instruction, which is used to describe the question asked by the user and is related to intent analysis, sentiment analysis, or negative sentiment analysis, generates an answer text for the question text, and outputs the answer text for the question text. The data storage device obtains the answer text for the question text output by the intent analysis model, and displays the answer text for the question text in the user interaction page.

[0072] Optionally, after detecting the problem scenario of the question text through the question analysis model, it also includes: when it is detected that the problem scenario of the question text is a common chat scenario, generating an answer text for the question text through a chat dialogue model, and displaying the answer text for the question text in a user interaction page.

[0073] Optionally, if the question scenario of the question text is a normal chat scenario, the question text is a text used to describe the questions asked by the user related to normal chat, and the answer text of the question text is a text used to describe the answers to the questions asked by the user related to normal chat described by the question text.

[0074] Optionally, a pre-trained chat dialogue model is provided in the data storage device. The pre-trained chat dialogue model may be a large-scale language model pre-trained by a technician and provided in the data storage device for generating an answer text for a question text describing a question related to ordinary chat inquired by a user according to a second text generation prompt instruction. The second text generation prompt instruction is a text for instructing the pre-trained chat dialogue model to detect and analyze the question text for describing the question related to ordinary chat inquired by the user, and generate an answer text for the question text. The second text generation prompt instruction contains the question text for describing the question related to ordinary chat inquired by the user, for which an answer text needs to be generated. The second text generation prompt instruction is input into the pre-trained chat dialogue model, and the pre-trained chat dialogue model will detect and analyze the question text for describing the question related to ordinary chat inquired by the user contained in the second text generation prompt instruction, generate an answer text for the question text, and output the answer text for the question text.

[0075] Optionally, an answer text to the question text is generated through a chat dialogue model, including: generating a second text generation prompt instruction corresponding to the question text based on the question text and a second text generation prompt instruction template; inputting the second text generation prompt instruction into a pre-trained chat dialogue model to obtain the answer text to the question text output by the chat dialogue model.

[0076] Optionally, the second text generation prompt instruction template is a second text generation prompt instruction that does not contain a question text for describing a question related to ordinary chat asked by a user and for which an answer text needs to be generated. The second text generation prompt instruction template contains a text filling position. The text filling position is a position for filling the question text for describing a question related to ordinary chat asked by a user and for which an answer text needs to be generated. After filling the question text for describing a question related to ordinary chat asked by a user and for which an answer text needs to be generated into the text filling position in the second text generation prompt instruction template, a second text generation prompt instruction corresponding to the question text for describing a question related to ordinary chat asked by a user and for which an answer text needs to be generated can be obtained. The second text generation prompt instruction corresponding to the question text for describing a question related to ordinary chat asked by a user and for which an answer text needs to be generated is a text for instructing a pre-trained chat dialogue model to detect and analyze the question text and generate an answer text for the question text. The second text generation prompt instruction template is stored in the data storage device.

[0077] Optionally, generating a second text generation prompt instruction corresponding to the question text according to the question text and the second text generation prompt instruction template includes: filling the question text into the text filling position in the second text generation prompt instruction template, thereby obtaining the second text generation prompt instruction corresponding to the question text. The second text generation prompt instruction corresponding to the question text is a text for instructing a pre-trained chat dialogue model to detect and analyze the question text used to describe a question asked by a user and related to ordinary chat, and generate an answer text to the question text.

[0078] Optionally, the data storage device inputs the second text generation prompt instruction corresponding to the question text into a pre-trained chat dialogue model. The pre-trained chat dialogue model detects and analyzes the question text contained in the second text generation prompt instruction for describing the question related to ordinary chat asked by the user, generates an answer text for the question text, and outputs the answer text for the question text. The data storage device obtains the answer text for the question text output by the chat dialogue model, and displays the answer text for the question text in the user interaction page.

[0079] The technical solution of the embodiment of the present invention is to detect the problem scenario of the problem text through the problem analysis model after obtaining the problem text input by the user; when it is detected that the problem scenario of the problem text is a data query scenario, the query description text of each target information and the data table range information of each target information are generated through the query task splitting model; then, the data query statement of each target information is generated according to the query description text of each target information through the data query model, and the business data related to each target information is obtained from the preset database according to the data query statement of each target information, and the query description text, data query statement and business data related to each target information of each target information are displayed on the user interaction page; finally, the query summary model is used to According to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, and the data query summary text of the question text is displayed in the user interaction page, which solves the problems of low query efficiency, high labor cost and time cost, and difficult to ensure accuracy of data query solutions in related technologies. Based on a large-scale language model and the question text entered by the user when it is necessary to query business data related to multiple target information in the business data table in the database, the business data related to each target information can be automatically queried in the business data table in the database, the business data related to each target information is obtained, and the business data related to each target information is provided to the user through the interactive page, the labor cost and time cost are low, and the accuracy and efficiency of the data query process are improved.

[0080] Embodiment 2

[0081] Figure 2 The flowchart of a data query method provided by the second embodiment of the present invention. The embodiment of the present invention can be combined with each optional solution in one or more of the above embodiments. Figure 2 As shown, the method includes:

[0082] Step 201: after obtaining the question text input by the user, generate a question analysis prompt instruction corresponding to the question text according to the question text and a question analysis prompt instruction template.

[0083] Step 202: input the question analysis prompt instruction into a pre-trained question analysis model to obtain the problem scenario of the question text output by the question analysis model: if the problem scenario of the question text is a data query scenario, execute step 203; if the problem scenario of the question text is a personnel intention analysis scenario, execute step 206; if the problem scenario of the question text is a normal chat scenario, execute step 207.

[0084] Step 203: Generate a query description text for each target information according to each target information in the question text and the data table range information of each target information through the query task splitting model.

[0085] Step 204: Generate data query statements for each target information based on the query description text of each target information through the data query model, and obtain business data related to each target information from a preset database based on the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page.

[0086] Step 205: Generate a data query summary text of the question text based on the query description text of each target information and the business data related to each target information through the query summary model, and display the data query summary text of the question text in the user interaction page.

[0087] Step 206: Generate an answer text for the question text through the intention analysis model, and display the answer text for the question text in the user interaction page.

[0088] Step 207: Generate an answer text to the question text through the chat dialogue model, and display the answer text to the question text in the user interaction page.

[0089] The technical solution of the embodiment of the present invention can automatically query the business data related to each target information in the business data table in the database based on the large-scale language model and the question text entered by the user when the user is in a scenario where it is necessary to query the business data related to multiple target information in the business data table in the database, obtain the business data related to each target information, and provide it to the user through an interactive page. It can generate corresponding answer text based on the large-scale language model and the question text entered by the user when the user is in a scenario where intent analysis, sentiment analysis or negative sentiment analysis is required, and provide it to the user through an interactive page. It can generate corresponding answer text based on the large-scale language model and the question text entered by the user when the user is in a scenario where ordinary chat is required, and provide it to the user through an interactive page.

[0090] Embodiment 3

[0091] Figure 3 This is a schematic diagram of the structure of a data query device provided in Embodiment 3 of the present invention. The device may be configured in an electronic device. Figure 3 As shown, the device includes: a scene detection module 301, a query task splitting module 302, a database query module 303 and a query summarizing module 304.

[0092] Among them, the scene detection module 301 is used to detect the problem scene of the question text through the problem analysis model after obtaining the question text input by the user; the query task splitting module 302 is used to generate the query description text of each target information through the query task splitting model according to the various target information in the question text and the data table range information of each target information when it is detected that the problem scene of the question text is a data query scene; the database query module 303 is used to generate the data query statement of each target information according to the query description text of each target information through the data query model, and obtain the business data related to each target information from the preset database according to the data query statement of each target information, and display the query description text, data query statement and business data related to each target information of each target information in the user interaction page; the query summary module 304 is used to generate the data query summary text of the question text according to the query description text of each target information and the business data related to each target information through the query summary model, and display the data query summary text of the question text in the user interaction page.

[0093] The technical solution of the embodiment of the present invention is to detect the problem scenario of the problem text through the problem analysis model after obtaining the problem text input by the user; when it is detected that the problem scenario of the problem text is a data query scenario, the query description text of each target information and the data table range information of each target information are generated through the query task splitting model; then, the data query statement of each target information is generated according to the query description text of each target information through the data query model, and the business data related to each target information is obtained from the preset database according to the data query statement of each target information, and the query description text, data query statement and business data related to each target information of each target information are displayed on the user interaction page; finally, the query summary model is used to According to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, and the data query summary text of the question text is displayed in the user interaction page, which solves the problems of low query efficiency, high labor cost and time cost, and difficult to ensure accuracy of data query solutions in related technologies. Based on a large-scale language model and the question text entered by the user when it is necessary to query business data related to multiple target information in the business data table in the database, the business data related to each target information can be automatically queried in the business data table in the database, the business data related to each target information is obtained, and the business data related to each target information is provided to the user through the interactive page, the labor cost and time cost are low, and the accuracy and efficiency of the data query process are improved.

[0094] In an optional implementation of an embodiment of the present invention, optionally, when the scene detection module 301 performs an operation of detecting the problem scenario of the question text through a question analysis model, it is specifically used to: generate a question analysis prompt instruction corresponding to the question text according to the question text and a question analysis prompt instruction template; input the question analysis prompt instruction into a pre-trained question analysis model to obtain the problem scenario of the question text output by the question analysis model; wherein the problem scenario of the question text is a data query scenario, a personnel intention analysis scenario or a normal chat scenario.

[0095] In an optional implementation of an embodiment of the present invention, optionally, when the query task splitting module 302 performs the operation of generating query description texts for respective target information according to respective target information in the question text and data table range information of respective target information through a query task splitting model, it is specifically used for: generating query task splitting prompt instructions corresponding to the question text according to the question text and a query task splitting prompt instruction template; inputting the query task splitting prompt instructions into a pre-trained query task splitting model, and obtaining query description texts for respective target information generated by the query task splitting model according to respective target information in the question text and data table range information of respective target information.

[0096] In an optional implementation of an embodiment of the present invention, optionally, when the database query module 303 executes the operation of generating data query statements for each target information according to the query description text of each target information through the data query model, it is specifically used to: generate prompt instruction templates according to the query description text and query statement of each target information, and generate prompt instructions for query statements corresponding to each target information; input the query statement generation prompt instructions corresponding to each target information into the pre-trained data query model, and obtain the data query statements for each target information output by the data query model.

[0097] In an optional implementation of an embodiment of the present invention, optionally, when the query summary module 304 executes the operation of generating a data query summary text of the question text according to the query description text of each target information and the business data related to each target information through the query summary model, it is specifically used to: generate a summary text generation prompt instruction corresponding to the question text according to the query description text of each target information, the business data related to each target information and a summary text generation prompt instruction template; input the summary text generation prompt instruction into a pre-trained query summary model to obtain the data query summary text of the question text generated according to the query description text of each target information and the business data related to each target information output by the query summary model.

[0098] In an optional implementation of an embodiment of the present invention, optionally, the data query device also includes: a first answer text generation module, which is used to generate an answer text for the question text through an intention analysis model when it is detected that the problem scenario of the question text is a personnel intention analysis scenario, and display the answer text of the question text in the user interaction page.

[0099] In an optional implementation of an embodiment of the present invention, optionally, the data query device also includes: a second answer text generation module, which is used to generate an answer text for the question text through a chat dialogue model when it is detected that the question scenario of the question text is a normal chat scenario, and display the answer text of the question text in the user interaction page.

[0100] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0101] The above-mentioned data query device can execute the data query method provided by any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the data query method.

[0102] Embodiment 4

[0103] Figure 4 The structure diagram of the electronic device 10 that can be used to implement the data query method of the embodiment of the present invention is shown. The components, their connections and relationships, and their functions shown herein are only examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0104] like Figure 4 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program executed by at least one processor, and the processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program built from the storage unit 18 to the random access memory (RAM) 13. In the random access memory (RAM) 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the read-only memory (ROM) 12 and the random access memory (RAM) 13 are connected to each other through a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0105] A number of components in the electronic device 10 are connected to an input / output (I / O) interface 15, including an input unit 16, such as a keyboard, a mouse, etc., an output unit 17, such as various types of displays, speakers, etc., a storage unit 18, such as a disk, an optical disk, etc., and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0106] The processor 11 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a data query method.

[0107] In some embodiments, the data query method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 10 via a read-only memory (ROM) 12 and / or a communication unit 19. When the computer program is built into a random access memory (RAM) 13 and executed by the processor 11, one or more steps of the data query method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to perform the data query method in any other appropriate manner (e.g., by means of firmware).

[0108] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0109] The computer program for implementing the data query method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0110] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0111] To provide interaction with a user, the systems and techniques described herein may be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0112] The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0113] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0114] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0115] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A data query method, characterized in that: include: After obtaining the question text input by the user, the question scenario of the question text is detected through the question analysis model; When it is detected that the problem scenario of the question text is a data query scenario, a query description text of each target information is generated according to each target information in the question text and data table range information of each target information through a query task splitting model; Generate data query statements for each target information based on the query description text of each target information through the data query model, obtain business data related to each target information from a preset database based on the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page; Through the query summary model, according to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, and the data query summary text of the question text is displayed in the user interaction page.

2. The data query method according to claim 1, characterized in that: Detecting the problem scenario of the problem text through the problem analysis model includes: Generate a question analysis prompt instruction corresponding to the question text according to the question text and the question analysis prompt instruction template; Inputting the problem analysis prompt instruction into a pre-trained problem analysis model to obtain the problem scenario of the problem text output by the problem analysis model; The problem scenario of the question text is a data query scenario, a personnel intention analysis scenario or a general chat scenario.

3. The data query method according to claim 1, characterized in that: Through the query task splitting model, according to each target information in the question text and the data table range information of each target information, a query description text of each target information is generated, including: Generate a query task splitting prompt instruction corresponding to the question text according to the question text and the query task splitting prompt instruction template; The query task splitting prompt instruction is input into a pre-trained query task splitting model to obtain the query description text of each target information generated by the query task splitting model according to each target information in the question text and the data table range information of each target information.

4. The data query method according to claim 1, characterized in that: Through the data query model, according to the query description text of each target information, the data query statement of each target information is generated, including: Generate prompt instruction templates according to the query description text and query statement of each target information respectively, and generate query statement generation prompt instructions corresponding to each target information; The query statement generation prompt instructions corresponding to each target information are respectively input into the pre-trained data query model to obtain the data query statement of each target information output by the data query model.

5. The data query method according to claim 1, characterized in that: Through the query summary model, according to the query description text of each target information and the business data related to each target information, a data query summary text of the question text is generated, including: Generate a summary text generation prompt instruction corresponding to the question text according to the query description text of each target information, the business data related to each target information and the summary text generation prompt instruction template; The summary text generation prompt instruction is input into a pre-trained query summary model to obtain a data query summary text of the question text generated by the query summary model based on the query description text of each target information and the business data related to each target information.

6. The data query method according to claim 2, characterized in that: After detecting the problem scenario of the problem text through the problem analysis model, the method further includes: When it is detected that the question scenario of the question text is a human intention analysis scenario, an answer text of the question text is generated through an intention analysis model, and the answer text of the question text is displayed in a user interaction page.

7. The data query method according to claim 2, characterized in that: After detecting the problem scenario of the problem text through the problem analysis model, the method further includes: When it is detected that the question scenario of the question text is a common chat scenario, an answer text of the question text is generated through a chat dialogue model, and the answer text of the question text is displayed in a user interaction page.

8. A data query device, characterized in that: include: A scene detection module is used to detect the problem scene of the question text through the question analysis model after obtaining the question text input by the user; A query task splitting module, for generating a query description text for each target information according to each target information in the question text and the data table range information of each target information through a query task splitting model when it is detected that the problem scenario of the question text is a data query scenario; The database query module is used to generate a data query statement for each target information according to the query description text of each target information through the data query model, and obtain the business data related to each target information from the preset database according to the data query statement of each target information, and display the query description text, data query statement and business data related to each target information on the user interaction page; The query summary module is used to generate a data query summary text of the question text based on the query description text of each target information and the business data related to each target information through a query summary model, and display the data query summary text of the question text in the user interaction page.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively coupled to the at least one processor; The memory stores a computer program executed by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the data query method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the data query method according to any one of claims 1 to 7 when executed.

Images