Data query method and system based on AI large model, terminal and medium

Abstract

The invention discloses a data query method and system based on an AI large model, a terminal and a medium, and the method comprises the steps: obtaining query request information, carrying out the intention recognition of the query request information, and obtaining a query intention; outputting query prompt information through an AI large model according to the query intention; and determining prompt type information of the query prompt information, and outputting a data query result based on the prompt type information. According to the method, the query intention can be recognized through the AI large model, then related follow-up operation is carried out according to the query intention, and then the data query result is returned, so that the use efficiency of the platform is improved, and a data basis is provided for follow-up decision making.

Description

Technical Field

[0001] The present invention relates to the field of AI big model technology, and in particular to a data query method, system, terminal and medium based on AI big model. Background Art

[0002] In the prior art, data query or information retrieval is basically implemented by manual input query, which requires much manual intervention and has low work efficiency. Especially in enterprise information platforms, data information is complex and diverse. If manual input query is still used, it cannot meet the needs of efficient work, and errors are easily caused by manual intervention.

[0003] Therefore, the prior art still needs to be improved and enhanced. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a data query method, system, terminal and medium based on an AI big model in view of the above-mentioned defects of the prior art. The technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present invention provides a data query method based on an AI big model, wherein the method comprises:

[0006] Acquire query request information, perform intent recognition on the query request information, and obtain query intent;

[0007] According to the query intent, query prompt information is output through the AI ​​big model;

[0008] Prompt type information of the query prompt information is determined, and a data query result is output based on the prompt type information.

[0009] In one implementation, the acquiring query request information, performing intent recognition on the query request information, and obtaining the query intent includes:

[0010] Acquire question information input by a user, and determine the query request information based on the question information;

[0011] The query request information is subjected to intent recognition through the AI ​​big model to obtain the query intent.

[0012] In one implementation, the performing intent recognition on the query request information by using an AI big model to obtain the query intent includes:

[0013] Parsing the query request information according to the AI ​​big model to obtain query keywords;

[0014] Rewrite the query keyword based on the AI ​​big model to obtain a rewritten query request;

[0015] Based on the rewritten query request, the query intent is output through the AI ​​big model.

[0016] In one implementation, the performing intent recognition on the query request information by using an AI big model to obtain the query intent includes:

[0017] The query keywords are filtered to remove illegal words or special characters.

[0018] In one implementation, outputting query prompt information through an AI big model according to the query intent includes:

[0019] Based on the query intent, calling the corresponding task agent;

[0020] The query intention is processed based on the task agent, and the corresponding vector library is called to output the query prompt information.

[0021] In one implementation, outputting the data query result based on the prompt type information includes:

[0022] If the prompt type information is an SQL prompt word, then generate a query SQL;

[0023] A data query is executed based on the query SQL to obtain a target data set, and the target data set is used as the data query result.

[0024] In one implementation, outputting the data query result based on the prompt type information includes:

[0025] If the prompt type information is question answer information, the question answer information is used as the data query result.

[0026] In a second aspect, an embodiment of the present invention further provides a data query system based on an AI big model, wherein the system comprises:

[0027] A query intent recognition module is used to obtain query request information, perform intent recognition on the query request information, and obtain query intent;

[0028] A query prompt output module, used to output query prompt information through an AI big model according to the query intent;

[0029] The query result output module is used to determine the prompt type information of the query prompt information and output the data query result based on the prompt type information.

[0030] In a third aspect, an embodiment of the present invention further provides a terminal, wherein the terminal includes a memory, a processor, and a data query program based on an AI big model stored in the memory and executable on the processor, and when the processor executes the data query program based on the AI ​​big model, the steps of the data query method based on the AI ​​big model of any one of the above-mentioned schemes are implemented.

[0031] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, wherein a data query program based on an AI big model is stored on the computer-readable storage medium, and when the data query program based on the AI ​​big model is executed by a processor, the steps of the data query method based on the AI ​​big model described in any one of the above-mentioned schemes are implemented.

[0032] Beneficial effects: Compared with the prior art, the present invention provides a data query method based on an AI big model. The present invention first obtains query request information, performs intent recognition on the query request information, and obtains query intent. Then, according to the query intent, query prompt information is output through the AI ​​big model. Finally, the prompt type information of the query prompt information is determined, and the data query result is output based on the prompt type information. The present invention can identify the query intent through the AI ​​big model, and then perform related subsequent operations according to the query intent, and then return the data query result, which improves the efficiency of the platform and provides a data basis for subsequent decision-making. The entire process of the present invention is realized automatically, without the need for human intervention, reducing the possibility of errors, improving work efficiency, and making user operations more convenient and the experience more friendly through intelligent interaction, thereby improving user stickiness. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A flowchart of a preferred embodiment of a data query method based on an AI big model provided in an embodiment of the present invention.

[0034] Figure 2 A schematic diagram of the application of the data query method based on the AI ​​big model provided in an embodiment of the present invention.

[0035] Figure 3 A schematic diagram of the architecture of a data query system based on an AI big model provided in an embodiment of the present invention.

[0036] Figure 4 A functional block diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and effect of the present invention clearer and more specific, the present invention is further described in detail with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0038] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations or steps, nor must they be executed in the order described. For example, some operations or steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0039] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0040] It should be understood that, in order to clearly describe the technical solutions of the embodiments of the present invention, in the embodiments of the present invention, words such as "first" and "second" are used to distinguish between identical or similar items with substantially identical functions and effects. For example, the first control information and the second control information are only used to distinguish different control information, and their order is not limited.

[0041] Those skilled in the art can understand that the words "first", "second", etc. do not limit the quantity and execution order, and the words "first", "second", etc. do not necessarily limit the differences.

[0042] It should be further understood that the term “and / or” used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0043] In order to solve the problems of the prior art, the present embodiment provides a data query method based on an AI big model. Based on the method of the present embodiment, the entire data query process is automatically implemented without manual intervention, which reduces the possibility of errors and improves work efficiency. Through intelligent interaction, user operation is more convenient, the experience is more friendly, and user stickiness is improved. In specific implementation, the present embodiment first obtains query request information, performs intent recognition on the query request information, and obtains the query intent. Then, according to the query intent, query prompt information is output through the AI ​​big model. Finally, the prompt type information of the query prompt information is determined, and the data query result is output based on the prompt type information. It can be seen that the present embodiment can identify the query intent through the AI ​​big model, and then perform relevant subsequent operations according to the query intent, and then return the data query result, which improves the efficiency of the platform and provides a data basis for subsequent decision-making.

[0044] The data query method based on the AI ​​big model of this embodiment can be applied to terminals, including intelligent product terminals such as computers, smart TVs, and mobile phones. Figure 1As shown in , the data query method based on the AI ​​big model in this embodiment includes the following steps:

[0045] Step S100: Acquire query request information, perform intent recognition on the query request information, and obtain query intent.

[0046] In this embodiment, the query request information may be a question raised by the user through the dialogue window of the platform, and the purpose of the query request information is to query a certain information or request a certain answer. When the AI ​​big model receives the query request information, it can perform intent recognition on the query request information, determine the query intent corresponding to the query request information, and the query intent reflects the user's request purpose.

[0047] In one implementation, this embodiment includes the following steps when determining the query intent corresponding to the query request information:

[0048] Step S101: obtaining question information input by a user, and determining the query request information based on the question information;

[0049] Step S102: parsing the query request information according to the AI ​​big model to obtain query keywords;

[0050] Step S103: rewrite the query keyword based on the AI ​​big model to obtain a rewritten query request;

[0051] Step S104: Based on the rewritten query request, the query intent is output through the AI ​​big model.

[0052] Combination Figure 2 As shown, a question-and-answer window is provided on the platform of the present embodiment. First, the question-and-answer window can be called to ask questions and obtain question information. In specific applications, the question information of the present embodiment can be text information entered by the user in the question-and-answer window, or it can be voice information entered by the user. Then, the query request information is determined based on the question information. Next, the present embodiment can call the "question-and-answer interface service" to connect to the AI ​​big model, and then analyze and process the query request information through the AI ​​big model. In the present embodiment, the query request information is first parsed by the AI ​​big model to obtain several query keywords. Then, the present embodiment inputs these query keywords into the filtering module to filter out some illegal words or special characters that affect the security of the system, so as to increase the accuracy of intent recognition in subsequent steps.

[0053] After filtering the query keywords, this embodiment can input the filtered query keywords into the query rewriting module of the AI ​​big model, and rewrite the query keywords by interacting with the query rewriting module of the AI ​​big model, so as to obtain the rewritten query, that is, to obtain the rewritten query request, so as to solve the problem that the user's colloquial language or description is vague and the intention cannot be accurately understood. Then, this embodiment interacts with the AI ​​big model to realize the intent recognition of the rewritten query request and obtain the corresponding query intent. Determining the query intent means knowing the user's operation intention, such as the intention to query personal performance information, or the intention to query weather conditions, etc.

[0054] Step S200: Output query prompt information through the AI ​​big model according to the query intention.

[0055] After the AI ​​big model recognizes the query intent, the AI ​​big model in this embodiment performs corresponding processing tasks according to the recognized query intent, and then outputs query prompt information, which can help users obtain data query results faster.

[0056] In one implementation, when outputting query prompt information, this embodiment includes the following steps:

[0057] Step S201: Based on the query intention, call the corresponding task agent;

[0058] Step S202: Process the query intention based on the task agent, call the corresponding vector library, and output the query prompt information.

[0059] Specifically, combined Figure 2 As shown in , in specific applications, after identifying the query intent, this embodiment can call the corresponding task agent (Agent) through the AI ​​big model, and use the task agent to process the user's query request information, call function_call (a public method for responding to requests) or call the vector library to match the user-related content, thereby obtaining the query prompt information (prompt). In this embodiment, the output query prompt information can be a SQL prompt word or a question answer information, and the question answer information is the answer to the question information raised by the user. In this embodiment, if the logic that needs to be processed by the user's query request information is relatively complex, it can be arranged through multiple Agent processes and multiple task modules to process user problems. Specifically, Figure 2 As shown in Figure 2The full-scenario agent in the system includes demand agent, reminder agent and task agent. Based on the arrangement of these multiple agents, it is convenient to process query request information more accurately based on the identified query intent, thereby obtaining more accurate query prompt information.

[0060] In one implementation, Figure 2 As shown, in this embodiment, parameters can be supplemented for multiple rounds of dialogue, so that the supplementary information can help the AI ​​large model to achieve more accurate and efficient processing. In practical applications, the supplementation of multiple rounds of dialogue can be achieved based on the user service interaction layer, that is, to achieve direct interaction between people and agents.

[0061] Step S300: Determine the prompt type information of the query prompt information, and output the data query result based on the prompt type information.

[0062] After the AI ​​big model outputs the query prompt information, this embodiment can output the corresponding data query result based on the prompt type information of the query prompt information. Specifically, if the prompt type information is an SQL prompt word, a query SQL is generated. Then, based on the query SQL, a data query is executed, data is assembled, and finally a target data set is obtained, and the target data set is used as the data query result. If the prompt type information is question answer information, there is no need to call the database, and the question answer information is directly used as the data query result.

[0063] Furthermore, this embodiment can also perform output filtering, perform data filtering operations such as permission data filtering or sensitive information filtering on the output data query results, and then call the model to summarize and analyze the data query results to generate a summary data set. In this embodiment, each user's dialogue information (i.e., question information) will be stored in the database, and there will also be a negative feedback operation where the answer is displayed on the interface, collecting question and answer data and forming historical dialogue data, which is convenient for subsequent question optimization and improves accuracy.

[0064] Based on the method of this embodiment, the query intent can be identified through the AI ​​big model, and then the relevant subsequent operations can be performed according to the query intent, and then the data query results can be returned, which improves the efficiency of the platform and provides a data basis for subsequent decision-making. For example, the user can enter "query the tasks I need to complete in September" through the client, and then call the system interface, match the user's query intent; after routing to the user task query Agent, it is matched that the user can get the result by querying the database, so the corresponding query SQL can be generated. Then, by executing the database query, the data set is obtained, for example, the user has 2 task data sets to be completed. Then, the model analyzes and summarizes according to the data set structure, generates the result data and returns it to the user. If the system interface is called, if the user's weather query intent is matched; then the function_call of the weather query is routed, and the function_call returns the weather information after calling the weather interface; the user finally obtains the weather information. Similarly, if the system interface is called, the user queries the enterprise system information, then the vector library is routed to obtain the enterprise knowledge information related to the system, and the content is finally returned to the user after the model is summarized. The entire process of this embodiment is implemented automatically without human intervention, which reduces the possibility of errors and improves work efficiency. Through intelligent interaction, user operation is more convenient, the experience is more friendly, and user stickiness is improved.

[0065] Based on the above embodiments, the present invention also provides a data query system based on AI big model, such as Figure 3 As shown, the system includes: a query intention recognition module 10, a query prompt output module 20 and a query result output module 30. Specifically, the query intention recognition module 10 is used to obtain query request information, perform intent recognition on the query request information, and obtain query intent. The query prompt output module 20 is used to output query prompt information through an AI big model according to the query intent. The query result output module 30 is used to determine the prompt type information of the query prompt information, and output the data query result based on the prompt type information.

[0066] In one implementation, the query intention identification module 10 includes:

[0067] A request information acquisition unit, used to acquire question information input by a user, and determine the query request information based on the question information;

[0068] The query intent determination unit is used to identify the intent of the query request information through an AI big model to obtain the query intent.

[0069] In one implementation, the query intent determination unit includes:

[0070] A keyword determination subunit, used to parse the query request information according to the AI ​​big model to obtain query keywords;

[0071] A query rewriting subunit, configured to rewrite the query keyword based on the AI ​​big model to obtain a rewritten query request;

[0072] The intention determination subunit is used to output the query intention through the AI ​​big model based on the rewritten query request.

[0073] In one implementation, the query intention determination unit further includes:

[0074] The filtering subunit is used to filter the query keywords and remove illegal words or special characters.

[0075] In one implementation, the query prompt output module 20 includes:

[0076] An agent calling unit, used to call a corresponding task agent based on the query intention;

[0077] A prompt output unit is used to process the query intention based on the task agent, call the corresponding vector library, and output the query prompt information.

[0078] In one implementation, the query result output module 30 includes:

[0079] A query SQL generating unit, configured to generate a query SQL if the prompt type information is an SQL prompt word;

[0080] The query SQL execution unit is used to execute data query based on the query SQL, obtain a target data set, and use the target data set as the data query result.

[0081] In one implementation, the query result output module 30 further includes:

[0082] A question answer determination unit is used to use the question answer information as the data query result if the prompt type information is question answer information.

[0083] The working principles of each module in the data query system based on the AI ​​big model of this embodiment are the same as the principles of each step in the above method embodiment, and will not be repeated here.

[0084] Each module in the above-mentioned data query system based on AI big model can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the terminal in the form of hardware, or can be stored in the memory in the terminal in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules.

[0085] Based on the above embodiment, the present invention further provides a terminal, the principle block diagram of the terminal can be as follows: Figure 4 The terminal may include one or more processors 100 ( Figure 4 Only one is shown in the figure), a memory 101 and a computer program 102 stored in the memory 101 and executable on one or more processors 100, for example, a sleep analysis program based on multi-sensor data. When one or more processors 100 execute the computer program 102, the various steps in the embodiment of the sleep analysis method based on multi-sensor data can be implemented. Alternatively, when one or more processors 100 execute the computer program 102, the functions of various modules / units in the embodiment of the sleep analysis system based on multi-sensor data can be implemented, which is not limited here.

[0086] In one embodiment, the processor 100 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0087] In one embodiment, the memory 101 may be an internal storage unit of an electronic device, such as a hard disk or memory of the electronic device. The memory 101 may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device. Further, the memory 101 may also include both an internal storage unit of the electronic device and an external storage device. The memory 101 is used to store computer programs and other programs and data required by the terminal. The memory 101 may also be used to temporarily store data that has been output or is to be output.

[0088] Those skilled in the art will understand that Figure 4 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the terminal to which the scheme of the present invention is applied. The specific terminal may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0089] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, operating database or other media used in the embodiments provided by the present invention may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double operational data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A data query method based on AI big model, characterized in that: The method comprises: Acquire query request information, perform intent recognition on the query request information, and obtain query intent; According to the query intent, query prompt information is output through the AI ​​big model; Prompt type information of the query prompt information is determined, and a data query result is output based on the prompt type information.

2. The data query method based on AI big model according to claim 1 is characterized in that: The acquiring of query request information, performing intent recognition on the query request information, and obtaining the query intent includes: Acquire question information input by a user, and determine the query request information based on the question information; The query request information is subjected to intent recognition through the AI ​​big model to obtain the query intent.

3. The data query method based on AI big model according to claim 1 is characterized in that: The performing intent recognition on the query request information by using the AI ​​big model to obtain the query intent includes: Parsing the query request information according to the AI ​​big model to obtain query keywords; Rewrite the query keyword based on the AI ​​big model to obtain a rewritten query request; Based on the rewritten query request, the query intent is output through the AI ​​big model.

4. The data query method based on AI big model according to claim 3 is characterized in that: The performing intent recognition on the query request information by using the AI ​​big model to obtain the query intent includes: The query keywords are filtered to remove illegal words or special characters.

5. The data query method based on AI big model according to claim 1 is characterized in that: Outputting query prompt information through the AI ​​big model according to the query intent includes: Based on the query intent, calling the corresponding task agent; The query intention is processed based on the task agent, and the corresponding vector library is called to output the query prompt information.

6. The data query method based on AI big model according to claim 1 is characterized in that: The outputting of the data query result based on the prompt type information includes: If the prompt type information is an SQL prompt word, then generate a query SQL; A data query is executed based on the query SQL to obtain a target data set, and the target data set is used as the data query result.

7. The data query method based on AI big model according to claim 6 is characterized in that: The outputting of the data query result based on the prompt type information includes: If the prompt type information is question answer information, the question answer information is used as the data query result.

8. A data query system based on AI big model, characterized in that: The system comprises: A query intent recognition module is used to obtain query request information, perform intent recognition on the query request information, and obtain query intent; A query prompt output module, used to output query prompt information through an AI big model according to the query intent; The query result output module is used to determine the prompt type information of the query prompt information and output the data query result based on the prompt type information.

9. A terminal, characterized in that: The terminal includes a memory, a processor, and a data query program based on an AI big model stored in the memory and executable on the processor. When the processor executes the data query program based on the AI ​​big model, the steps of the data query method based on the AI ​​big model as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a data query program based on the AI ​​big model. When the data query program based on the AI ​​big model is executed by the processor, the steps of the data query method based on the AI ​​big model as described in any one of claims 1 to 7 are implemented.

Images