Data analysis method, device, medium and equipment

Abstract

The present application provides a data analysis method, apparatus, medium and device, which relates to the field of data processing technology, including: obtaining input information determined by a user according to data to be queried; inputting the original text into an intent classification model to obtain a first intent; obtaining a second intent list L according to at least one fixed query label corresponding to each original label and each preset intent; if L 1 Same as the first intention, then L 1 Determine the target intent; obtain the data analysis result corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent, and the large language model. The final analysis result obtained by this application based on the input information and the target intent is also a result that is closer to the real needs of the user, improves the accuracy of data query and analysis, and avoids multiple interactions with the user as much as possible, thereby improving the efficiency of data query and analysis.

Description

Background Art

[0002] With the advent of the big data era, enterprises and organizations have an increasing demand for data management and analysis. As an important tool for data analysis and decision support, the indicator query system can extract key information from massive data based on preset indicators, providing an important basis for business decision-making.

[0003] Existing query systems based on preset indicators have been widely used. These systems usually allow users to query based on preset indicators (such as sales, number of users, conversion rate, etc.) to obtain statistical and analytical results of relevant data. However, in practical applications, these systems face a significant problem: users do not have a deep understanding of the indicators, resulting in inaccurate query results. Summary of the invention

[0004] In response to the above technical problems, the present application provides a data analysis method, device, medium and equipment to at least partially solve the problems existing in the prior art.

[0005] In a first aspect of the present application, a data analysis method is provided, the method comprising:

[0006] S100, obtaining input information determined by a user according to the data to be queried; wherein the input information includes original text and at least one original tag;

[0007] S200, inputting the original text into an intent classification model to obtain a first intent; wherein the first intent is any one of a plurality of preset intents corresponding to the intent classification model; each preset intent has at least one corresponding fixed query tag;

[0008] S300, obtaining a second intent list L=(L 1 , L 2 , …, L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =G i / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 ;

[0009] S400, if L 1 Same as the first intention, then L 1 Identify as target intention;

[0010] S500, obtaining data analysis results corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent, and a large language model.

[0011] In a second aspect of the present application, a data analysis device is provided, the device comprising:

[0012] An information acquisition unit, used to acquire input information determined by a user according to the data to be queried; wherein the input information includes original text and at least one original label;

[0013] A first intention acquisition unit is used to input the original text into the intention classification model to obtain a first intention; wherein the first intention is any one of a plurality of preset intentions corresponding to the intention classification model; each preset intention has at least one corresponding fixed query tag;

[0014] The second intent acquisition unit is configured to obtain a second intent list L=(L 1 , L 2 , …, L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =G i / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 ;

[0015] Target intention determination unit, used if L 1 Same as the first intention, then L 1 Identify as target intention;

[0016] A result generating unit is used to obtain a data analysis result corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent and the large language model.

[0017] In a third aspect of the present application, a non-transitory computer-readable storage medium is provided, in which at least one instruction or at least one program is stored, and the at least one instruction or at least one program is loaded and executed by a processor to implement the aforementioned data analysis method.

[0018] In a fourth aspect of the present application, an electronic device is provided, comprising a processor and the above-mentioned non-transitory computer-readable storage medium.

[0019] This application has at least the following beneficial effects:

[0020] In the data analysis method provided by the present application, the user can input text information independently while inputting a label, so as to implement the explanation of the information of the input label. Then, the corresponding first intention is obtained according to the input original text; and because there is more than one original label, each preset intention also has multiple fixed query labels, so a list of second intentions arranged in order of size is obtained, and the fixed label consistency ratio corresponding to each second intention in the second intention list is greater than the preset consistency ratio. That is, the higher the ranking of the second intention, the higher the comprehensive matching degree with all the original labels. Then, L 1 is the second intention with the highest comprehensive matching degree with all original tags. If L 1 If the original label is the same as the first intention, it means that the original label selected by the user is more accurate. Finally, the data analysis result corresponding to the data to be queried is obtained according to the input information, each fixed query label corresponding to the target intention, and the large language model. In this application, since the original text is a natural language, it is easier to be close to the user's real intention than the preset label (original label). Therefore, this application performs intent recognition through the original text input by the user to reversely determine the accuracy of the original label. If the second intention determined according to the original label has a higher comprehensive match with the first intention, Figure 1 If the result is consistent, it means that at least one of the original labels selected by the user is relatively accurate and can more appropriately display the user's true intention. Therefore, the final analysis result obtained based on the input information and target intention is also closer to the user's real needs, which improves the accuracy of data query and analysis, avoids multiple interactions with the user as much as possible, and improves the efficiency of data query and analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1A flowchart of a data analysis method provided in an embodiment of the present application;

[0023] Figure 2 This is a structural block diagram of the data analysis device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0025] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application 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 interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0026] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0027] Please refer to Figure 1 As shown, an embodiment of the present application provides a data analysis method, characterized in that the method includes:

[0028] S100, obtaining input information determined by a user according to data to be queried; wherein the input information includes original text and at least one original label.

[0029] Specifically, the user selects several original tags according to the data to be queried in the tag selection bar of the operation interface, and enters the original text according to the data to be queried. The input original text can be used to describe the user's query purpose, or to supplement the query purpose. As an example: the original tags can be flight volume, Air China. The original text can be: I want to query the flight volume of Air China between October 1 and October 7, 2024.

[0030] S200, inputting the original text into an intent classification model to obtain a first intent; wherein the first intent is any one of several preset intents corresponding to the intent classification model; and each preset intent has at least one corresponding fixed query tag.

[0031] Specifically, the input of the intent classification model is any preset intent included in several preset intents, and the input is the original text. And each preset intent has at least one corresponding fixed query label. Here, the fixed query label is a non-variable label. As an example, the fixed query label can be: domestic, international, Air China, and China Eastern Airlines. On the contrary, relative to the fixed query label, there is also a non-fixed query label. As an example, the non-fixed query label can be a time query label, and the time query label can be extracted based on the original text. Since the query result corresponding to the fixed query label is unique, and due to the variability of the time query label, for the same intent, the corresponding time period is variable. Therefore, each preset intent not only has a corresponding fixed query label, but also has a corresponding non-fixed query label.

[0032] S300, obtaining a second intent list L=(L 1 , L 2 , …, L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =G i / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 .

[0033] Specifically, if the fixed label consistency ratio of a second intent is higher, it means that the comprehensive matching degree with all original labels is higher, and the corresponding second intent is more consistent with the real intent corresponding to all original labels. In this embodiment, several second intents with a consistency ratio greater than the preset ratio are retained.

[0034] S400, if L 1 Same as the first intention, then L 1 Identify as target intent.

[0035] Specifically, if L 1 The second intention is the same as the first intention, that is, the second intention with a higher comprehensive matching degree with all original tags is the same as the first intention, which means that the original tag selected by the user is more accurate and can more appropriately show the user's true intention.

[0036] S500, obtaining data analysis results corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent, and a large language model.

[0037] Specifically, the step S500 includes:

[0038] S510: Generate a time query tag according to the original text and at least one original tag.

[0039] Here, for the same intent, the corresponding time period is variable. Therefore, each preset intent not only has a corresponding fixed query tag, but also has a corresponding non-fixed query tag, such as a time query tag. After determining the intent according to the fixed query tag, the corresponding non-fixed query tag is supplemented according to the time query tag. As an example: the time query tag can be between October 1 and October 7, 2024.

[0040] S520: Obtain a data query result corresponding to the data to be queried according to the time query tag, each fixed query tag corresponding to the target intent, and a preset database.

[0041] Here, according to the time query tag and each fixed query tag corresponding to the target intent, a query is performed in a preset database to obtain a data query result corresponding to the data to be queried, wherein the data query result may be statistical data or original data.

[0042] S530: Input the data query result into a large language model to obtain a data analysis result corresponding to the data to be queried.

[0043] Here, the big prediction model can analyze the input data query results, including data analysis, change trends, reflected problems, and corresponding industry conditions. As an example: the data query results are: domestic normal flights in this period: 3674; semi-international normal flights in this period: 283; domestic normal flights in the same period last year: 3869; semi-international normal flights in the same period last year: 187; pure domestic year-on-year data change: -0.05%; semi-international year-on-year data change: -0.51%; pure domestic year-on-year difference: -195; semi-international year-on-year difference: 96; The data analysis results obtained by the big language model are: In the latest statistics, the number of normal flights in the civil aviation industry showed a certain trend of change. Specifically, in the field of pure domestic routes, the number of normal flights in this period is 3674, which is 195 less than 3869 in the same period last year, a year-on-year decrease of 0.05%, which shows that in the operation of domestic routes, although the number of flights remains at a high level, it has declined slightly compared with last year. On semi-international routes, that is, routes between domestic and international destinations, the number of normal flights in this period reached 283, an increase of 96 flights compared with 187 flights in the same period last year, a year-on-year increase of 0.51%. This data change shows that in the operation of semi-international routes, the civil aviation industry has shown a relatively obvious growth trend, and the number of flights has increased significantly, a year-on-year increase of 0.51%. This data change shows that in the operation of international routes, the civil aviation industry has shown a relatively obvious growth trend, and the number of flights has increased significantly. Through comparative analysis with the same period last year, we can more clearly understand the operating status of the civil aviation industry on different route types, and whether it is showing an upward or downward trend. This comparison helps civil aviation companies to adjust their strategies in a timely manner, respond to market changes, and ensure operational efficiency and passenger satisfaction.

[0044] In an exemplary embodiment of the present application, after step S300, the method further includes:

[0045] S600, if L 1 Different from the first intention, and L except L 1 If any second intention other than 1 and generating a first corrected text using a text generation model.

[0046] Specifically, if L 1 Different from the first intention, and L except L 1 If any second intent other than L is the same as the first intent, it means that the most suitable second intent corresponding to all the original tags selected by the user is different from the first intent corresponding to the original text input by the user, but L 1Any second intent other than L is the same as the first intent, that is, some of the original tags selected by the user do not match the first intent corresponding to the original text input by the user. 1 and the text generation model reversely generates a first corrected text.

[0047] S700, outputting the first corrected text to a user operation interface; wherein the user operation interface is used to display an original text label, the first corrected text label, and non-original text and the first corrected text label.

[0048] Specifically, the first corrected text is output to the user operation interface for user operation. At this time, the operation interface has three optional options: original text label, first corrected text label, and non-original text and first corrected text label (neither).

[0049] S800: In response to detecting a click operation on the first correction text label, 1 Determine it as the target intention and jump to step S500.

[0050] Specifically, if the user selects the first correction text label, it means that the original text label input by the user is partially incorrect, and the original label selected by the user is closest to the user's true intention. 1 For the target intention. And conduct subsequent data query and data analysis.

[0051] In an exemplary embodiment of the present application, after step S700, the method further includes:

[0052] S1400, in response to detecting a click operation on the original text label, determining the first intention as the target intention and jumping to step S500.

[0053] Specifically, if the user selects the original text label, it means that the original text label input by the user is closer to the user's real intention. At this time, the first intention is determined as the target intention, and subsequent data query and data analysis are performed.

[0054] In this embodiment, when L 1 Different from the first intention, and L except L 1 Any second intent other than L is the same as the first intent, that is, some of the original tags selected by the user do not match the first intent corresponding to the original text input by the user. At this time, according to L 1The first corrected text is generated in reverse, because the first corrected text generated in reverse is natural language, which is convenient for users to judge. If the user selects the original text label, it means that the original text label input by the user is closer to the user's true intention. On the contrary, if the user selects the first corrected text label, it means that the original text label input by the user is partially incorrect, and the original label selected by the user is closest to the user's true intention. Then determine the corresponding target intention, and perform subsequent data query and data analysis. In this embodiment, the text generation model is used to generate the corrected text in reverse, and the system's confidence in the original text and original label is determined based on the user's feedback, so that the final data query results and data analysis results are closer to the user's true intention.

[0055] In an exemplary embodiment of the present application, after step S700, the method further includes:

[0056] S900, in response to a click operation on a non-original text and a corrected text label, a third intent is obtained based on the original text, each of the original labels and an intent classification model; wherein the original text is used as input and each of the original labels is used as an input prompt; and the third intent is any one of several preset intents corresponding to the intent classification model.

[0057] S1000, determine the third intention as the target intention, and jump to step S500.

[0058] In this embodiment, the user has neither selected the original text nor the corrected text, which means that at this time, the original text and the original label may have some content that is not close to the user's true intention. At this time, in order to improve the accuracy of user intention recognition, the original text is used as input and each of the original labels is used as an input prompt to obtain a third intention, which combines the content contained in the original text and the original label. 1 , closer to the user’s real intention.

[0059] In an exemplary embodiment of the present application, after step S900, the method further includes:

[0060] S1100, generating a second corrected text according to the third intention and the text generation model;

[0061] S1200, outputting the second corrected text to a user operation interface; wherein the user operation interface is used to display the original text label, the second corrected text label, and the non-original text and corrected text label;

[0062] S1300, in response to a click operation on the second correction text, the third intention is determined as the target intention, and the process jumps to step S500.

[0063] In this embodiment, in order to correct the third intention, a second correction text is reversely generated according to the third intention for the user to choose. If the user chooses the second correction text, it means that the third intention is closer to the user's true intention. At this time, the third intention is determined as the target intention, and subsequent data query and data analysis are performed. If the user selects the non-original text and correction text labels, the relevant information of the user's true intention can be obtained through multiple rounds of interaction to improve the authenticity of intent recognition.

[0064] In an exemplary embodiment of the present application, after step S300, the method further includes:

[0065] S1500: If the first intention is different from any second intention, the first intention is determined as the target intention, and the process jumps to step S500.

[0066] In this embodiment, if the first intent is different from any second intent, it means that the original label selection may deviate greatly. As a result, any corresponding second intent is different from the first intent. In this case, the first intent is determined as the target intent, and subsequent data query and data analysis are performed. Multiple interactions with the user are avoided, and the efficiency of data query and analysis is improved.

[0067] Please refer to Figure 2 As shown, an embodiment of the present application provides a data analysis device 100, the device comprising:

[0068] The information acquisition unit is used to acquire input information determined by the user according to the data to be queried; wherein the input information includes original text and at least one original label.

[0069] The first intention acquisition unit is used to input the original text into the intention classification model to obtain a first intention; wherein the first intention is any preset intention among several preset intentions corresponding to the intention classification model; each preset intention has at least one corresponding fixed query tag.

[0070] The second intent acquisition unit is configured to obtain a second intent list L=(L 1 , L 2 , …, L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =Gi / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 .

[0071] Target intention determination unit, used if L 1 Same as the first intention, then L 1 Identify as target intent.

[0072] A result generating unit is used to obtain a data analysis result corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent and the large language model.

[0073] In an exemplary embodiment of the present application, an electronic device capable of implementing the above method is also provided.

[0074] Those skilled in the art will appreciate that various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "system" herein.

[0075] The electronic device according to this embodiment of the present application is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0076] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the at least one processor mentioned above, the at least one storage device mentioned above, and a bus connecting different system components (including storage devices and processors).

[0077] The storage stores program codes, which can be executed by the processor, so that the processor executes the steps described in the above “Exemplary Method” section of this specification according to various exemplary embodiments of the present application.

[0078] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read only memory (ROM).

[0079] The storage may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules and program data, each of which or some combination may include the implementation of a network environment.

[0080] The bus may represent one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

[0081] The electronic device can also communicate with one or more external devices (such as keyboards, pointing devices, Bluetooth devices, etc.), and can also communicate with one or more devices that enable users to interact with the electronic device, and / or communicate with any device (such as routers, modems, etc.) that enables the electronic device to communicate with one or more other computing devices. This communication can be carried out through an input / output (I / O) interface. In addition, the electronic device can also communicate with one or more networks (such as local area networks (LANs), wide area networks (WANs) and / or public networks, such as the Internet) through a network adapter. As shown in the figure, the network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0082] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software or by combining software with necessary hardware. Therefore, the technical solution according to the implementation methods of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation methods of the present application.

[0083] In an exemplary embodiment of the present application, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present application can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present application described in the above "Exemplary Method" section of the present specification.

[0084] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable 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 above.

[0085] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0086] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0087] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0088] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0089] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0090] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A data analysis method, characterized in that: The method comprises: S100, obtaining input information determined by a user according to the data to be queried; wherein the input information includes original text and at least one original tag; S200, inputting the original text into an intent classification model to obtain a first intent; wherein the first intent is any one of a plurality of preset intents corresponding to the intent classification model; each preset intent has at least one corresponding fixed query tag; S300, obtaining a second intent list L=(L1, L2, ..., L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =G i / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 ; S400, if L1 is the same as the first intention, determining L1 as the target intention; S500, obtaining data analysis results corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent, and a large language model.

2. The data analysis method according to claim 1, characterized in that: After step S300, the method further includes: S600, if L1 is different from the first intention, and any second intention in L except L1 is the same as the first intention, generating a first corrected text according to L1 and a text generation model; S700, outputting the first corrected text to a user operation interface; wherein the user operation interface is used to display the original text label, the first corrected text label, and the non-original text and the first corrected text label; S800, in response to detecting a click operation on the first correction text label, L1 is determined as the target intention, and the process jumps to step S500.

3. The data analysis method according to claim 2, characterized in that: After step S700, the method further includes: S900, in response to a click operation on a non-original text and a corrected text label, a third intent is obtained according to the original text, each of the original labels and an intent classification model; wherein the original text is used as an input, each of the original labels is used as an input prompt; and the third intent is any one of a plurality of preset intents corresponding to the intent classification model; S1000, determine the third intention as the target intention, and jump to step S500.

4. The data analysis method according to claim 3, characterized in that: After step S900, the method further includes: S1100, generating a second corrected text according to the third intention and the text generation model; S1200, outputting the second corrected text to a user operation interface; wherein the user operation interface is used to display the original text label, the second corrected text label, and the non-original text and corrected text label; S1300, in response to a click operation on the second correction text, the third intention is determined as the target intention, and the process jumps to step S500.

5. The data analysis method according to claim 1, characterized in that: Step S500 includes: S510, generating a time query tag according to the original text and at least one original tag; S520, obtaining a data query result corresponding to the data to be queried according to the time query tag, each fixed query tag corresponding to the target intent, and a preset database; S530: Input the data query result into a large language model to obtain a data analysis result corresponding to the data to be queried.

6. The data analysis method according to claim 2, characterized in that: After step S700, the method further includes: S1400, in response to detecting a click operation on the original text label, determining the first intention as the target intention and jumping to step S500.

7. The data analysis method according to claim 2, characterized in that: After step S300, the method further includes: S1500: If the first intention is different from any second intention, the first intention is determined as the target intention, and the process jumps to step S500.

8. A data analysis device, characterized in that: The device comprises: An information acquisition unit, used to acquire input information determined by a user according to the data to be queried; wherein the input information includes original text and at least one original label; A first intention acquisition unit is used to input the original text into an intention classification model to obtain a first intention; wherein the first intention is any one of a plurality of preset intentions corresponding to the intention classification model; each preset intention has at least one corresponding fixed query tag; The second intent acquisition unit is configured to obtain a second intent list L=(L1, L2, ..., L i , …, L n ); i = 1, 2, ..., n; wherein n is the number of the second intention; L i is the i-th second intention; LB i ≥YB;LB i For L i Corresponding fixed label consistency ratio: LB i =G i / a;G i is the number of original tags that are identical to the fixed query tag corresponding to the i-th second intent; a is the total number of original tags; YB is the preset consistency ratio; and LB i ≥LB i+1 ; a target intention determining unit, configured to determine L1 as the target intention if L1 is the same as the first intention; A result generating unit is used to obtain a data analysis result corresponding to the data to be queried according to the input information, each fixed query tag corresponding to the target intent and the large language model.

9. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the method as claimed in any one of claims 1 to 7.

10. An electronic device, characterized in that: The invention comprises a processor and the non-transitory computer-readable storage medium as claimed in claim 9.

Images