Operation business data query analysis method and system based on large model, and electronic equipment

Abstract

The invention relates to the technical field of data processing, and discloses an operation business data query analysis method and system based on a large model and electronic equipment, and the method comprises the following steps: operating the large model to execute the following steps: collecting and analyzing an instruction to obtain query features and analysis features; obtaining a corresponding query result and an analysis result; corresponding query feedback and analysis feedback are obtained; and updating the corresponding query rule and analysis rule. The electronic equipment and the system correspond to the method. According to the method, technical support is provided for operation business data query analysis through the large model, the intelligence of query analysis is improved, data reference for query analysis is provided through the query features and the analysis features, the query result and the analysis result are automatically generated under the guidance of the query rules and the analysis rules, the interpretability of query analysis is improved, and the user experience is improved. And the applicability of the large model is improved by updating the query parameters and the associated parameters. And finally, operation business data query analysis based on the large model is realized.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and specifically to a method, system and electronic equipment for querying and analyzing operational business data based on a large model. Background Art

[0002] The rapid development of information technology has led to a rapid increase in the amount of data, far exceeding the level that humans can handle on their own. As the most intuitive data reflecting the current operating status, it is necessary to make operational business data easier to query and analyze, that is, to improve its interpretability.

[0003] Chinese patent number CN202311693760.0 discloses a production operation data analysis method, system, storage medium, and electronic device. The production operation data analysis method includes: obtaining production operation data and modeling the production operation data: creating a data table based on the data model obtained by modeling; when receiving a natural language query instruction, querying the data table according to the natural language query instruction to obtain query data; performing structured analysis on the query data to obtain structured data, and generating an insight report based on the structured data; visually displaying the structured data and the insight report. However, the invention only focuses on querying, and the insight report is generally good at the results of data analysis.

[0004] With the development of machine learning technology, it is necessary to study its application in operational business data query and analysis.

[0005] In summary, the existing technology urgently needs a technical solution for querying and analyzing operational business data based on a large model. Summary of the invention

[0006] The purpose of this application is to provide a method, system and electronic device for querying and analyzing operational business data based on a large model to solve the technical problems raised in the above-mentioned background technology.

[0007] To achieve the above objectives, this application discloses the following technical solutions:

[0008] In a first aspect, the present application discloses a method for querying and analyzing operational business data based on a large model, the method comprising:

[0009] S1: Run a preset large model, and the large model executes steps S2 to S5;

[0010] S2: Collect and parse instructions to obtain query features and analysis features;

[0011] S3: using the query feature and the analysis feature to obtain corresponding query results and analysis results;

[0012] S4: using the query result and the analysis result, obtaining corresponding query feedback and analysis feedback;

[0013] S5: using the query feedback and the analysis feedback, updating corresponding query rules and analysis rules;

[0014] The big model stores the preset query rules and the preset analysis rules, the query rules are used to obtain the query results based on the queried features, the analysis rules are used to obtain the analysis results based on the analysis features, the query features are used to characterize the query intent of the instructions, the analysis features are used to characterize the analysis intent of the instructions, the query results and the analysis results are used by the administrator to manage and operate business data, and the query feedback and the analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and the analysis results.

[0015] Preferably, the generation process of the query feature specifically includes:

[0016] The instructions are parsed using natural language technology, and the language of the query intent related to the operational business data in the instructions is defined as the query feature.

[0017] Preferably, the process of generating the analysis feature specifically includes:

[0018] The instructions are parsed using natural language technology, and the language of the analysis intent related to the operational business data in the instructions is defined as the analysis feature.

[0019] Preferably, the query rule includes a query table and query parameters;

[0020] The query table includes a root, a first query column, and a second query column;

[0021] The process of querying using the query table specifically includes:

[0022] Using the query feature to obtain the corresponding root word;

[0023] Using the root word to obtain the corresponding first query column, the first query column storing operational business data corresponding to the query feature;

[0024] The corresponding second query column is obtained by using the root word and the query parameter, and the second query column stores the operational business data corresponding to the query feature.

[0025] Preferably, the analysis rules include a first analysis mechanism and a second analysis mechanism;

[0026] The first analysis mechanism uses the analysis feature and the first query column to perform a self-analysis corresponding to the query feature;

[0027] The second analysis mechanism uses the analysis feature, the second query column and an association parameter to perform an association analysis corresponding to the query feature; the association parameter is used to configure the analysis association degree.

[0028] Preferably, the generation process of the query result and the query feedback specifically includes:

[0029] A1: Acquire the query feature and call the query rule;

[0030] A2: using the query feature obtained in step A1 to obtain the corresponding word root;

[0031] A3: Using the root word to obtain the corresponding first query column;

[0032] A4: Obtain the corresponding second query column using the root word and the query parameter;

[0033] A5: define the first query column obtained in step A3 and the second query column obtained in step A4 as the query result;

[0034] A6: Collect the feedback content input by the administrator after receiving the query result to obtain the query feedback.

[0035] Preferably, the generation process of the analysis result and the analysis feedback specifically includes:

[0036] B1: Acquire the analysis feature and call the analysis rule;

[0037] B2: using the analysis feature obtained in step B1, the first analysis mechanism and the first query column to perform the self-analysis corresponding to the query feature;

[0038] B3: performing association analysis corresponding to the query feature using the analysis feature obtained in step B1, the second analysis mechanism, the second query column and the association parameter;

[0039] B4: defining the result of the self-analysis in step B2 and the result of the association analysis in step B3 as the analysis result;

[0040] B5: Collecting feedback content input by the administrator after receiving the analysis result to obtain the analysis feedback.

[0041] Preferably, the query rules and analysis rules corresponding to the update specifically include:

[0042] The query parameters and the association parameters are updated based on the query feedback and the analysis feedback.

[0043] In a second aspect, the present application discloses a big model-based operational business data query and analysis system, the system is applicable to the big model-based operational business data query and analysis method as described above, and the system comprises:

[0044] A large model module, wherein the large model module is configured to store and run a feature unit, a result unit, a feedback unit, and an update unit;

[0045] The feature unit is configured to: collect and parse instructions to obtain query features and analysis features;

[0046] The result unit is configured to: obtain corresponding query results and analysis results using the query characteristics and the analysis characteristics;

[0047] The feedback unit is configured to: obtain corresponding query feedback and analysis feedback using the query result and the analysis result;

[0048] The updating unit is configured to: update corresponding query rules and analysis rules using the query feedback and the analysis feedback;

[0049] The big model stores the preset query rules and the preset analysis rules, the query rules are used to obtain the query results based on the queried features, the analysis rules are used to obtain the analysis results based on the analysis features, the query features are used to characterize the query intent of the instructions, the analysis features are used to characterize the analysis intent of the instructions, the query results and the analysis results are used by the administrator to manage and operate business data, and the query feedback and the analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and the analysis results.

[0050] In a third aspect, the present application discloses an electronic device, comprising: at least one processor and at least one memory, wherein the memory is communicatively connected to the processor; the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, the operational business data query and analysis method based on a large model as described above is implemented.

[0051] Beneficial effects: The operational business data query and analysis method, system and electronic device based on a big model of the present application provide technical support for operational business data query and analysis through a big model, improve the intelligence of the query and analysis, provide data reference for query and analysis through query features and analysis features, automatically generate query results and analysis results under the guidance of query rules and analysis rules, improve the interpretability of query analysis, and improve the applicability of the big model by updating query parameters and associated parameters, thereby realizing operational business data query and analysis based on a big model. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 A flowchart of a method for querying and analyzing operational business data based on a large model provided in an embodiment of the present application;

[0054] Figure 2 A structural block diagram of the large-model-based operational business data query and analysis system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0055] The technical solutions in the embodiments of the present application are described clearly and completely below. 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 ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0056] In this article, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, the elements defined by the sentence "comprising..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0057] In a first aspect, this embodiment discloses Figure 1 A method for querying and analyzing operational business data based on a large model is shown, the method comprising:

[0058] S1: Run the preset big model, and the big model executes steps S2 to S5. It should be noted that this embodiment uses existing machine learning technology and natural language processing technology to build a big model, aiming to use the big model to query and analyze operational business data;

[0059] S2: Collect and parse instructions to obtain query features and analysis features;

[0060] S3: using the query features and analysis features, obtaining corresponding query results and analysis results; it can be understood that the corresponding query results and analysis results obtained here are obtained by the large model calling the query rules and analysis rules;

[0061] S4: using the query results and analysis results, obtaining corresponding query feedback and analysis feedback; it can be understood that obtaining the corresponding query feedback and analysis feedback here is obtained by collecting the corresponding feedback content by the large model;

[0062] S5: using the query feedback and analysis feedback, updating the corresponding query rules and analysis rules;

[0063] The big model stores preset query rules and preset analysis rules. The query rules are used to obtain query results based on the queried features, and the analysis rules are used to obtain analysis results based on the analysis features. The query features are used to characterize the query intent of the instructions, and the analysis features are used to characterize the analysis intent of the instructions. The query results and analysis results are used by administrators to manage and operate business data. The query feedback and analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and analysis results.

[0064] Based on the above, this embodiment provides necessary technical support for query analysis through the big model, improves the intelligence of query analysis, obtains query features and analysis features by parsing instructions, provides a data basis for query analysis, provides data reference for query results and analysis results through query rules and analysis rules, provides a data basis for updating query rules and analysis rules through query feedback and analysis feedback, optimizes the big model by updating query rules and analysis rules, thereby improving the applicability of query analysis, provides a data basis for administrators to manage query results and analysis results through the obtained query results and analysis results, and realizes operational business data query and analysis based on the big model.

[0065] Specifically, the query feature generation process includes:

[0066] The instructions are parsed using natural language technology, and the language in the instructions involving the query intent of operational business data is defined as query features.

[0067] It should be noted that the instructions of this embodiment may be, but are not limited to, query text input by an administrator, which is obtainable by technical personnel in this field based on well-known experience, such as "query sales in the past three days", "query and analyze sales trends in the past month", etc., and the existing natural language processing technology is used to capture query features and analysis features. For example, if the language of the query intent involving operational business data of "query sales in the past three days" is "three days" and "sales", then the query features are "three days" and "sales", thereby providing an accurate data basis for query analysis.

[0068] Based on the above, this embodiment converts the language of the query intention related to the operational business data in the instruction into query features, thereby providing an accurate data basis for querying the large model.

[0069] Specifically, it is characterized in that the generation process of the analysis feature specifically includes:

[0070] The instructions are parsed using natural language technology, and the language in the instructions involving the analytical intent of the operational business data is defined as analytical features.

[0071] In a simple example, existing natural language processing technology is used to capture query features and analysis features. For example, the language of the query intent involving operational business data of "query and analyze the sales trend in the past month" is "one month" and "sales", and the query features are "one month" and "sales". The language of the analysis intent involving operational business data is "sales trend", and the analysis feature is "sales trend".

[0072] Based on the above, this embodiment converts the language of the analysis intention related to the operational business data in the instruction into analysis features, thereby providing an accurate data basis for the large model to perform analysis.

[0073] Specifically, the query rule includes a query table and query parameters;

[0074] The query table includes a root, a first query column, and a second query column;

[0075] The process of querying using the query table specifically includes:

[0076] Use query features to get corresponding word roots;

[0077] Using the root word, a corresponding first query column is obtained, where the first query column stores operational business data corresponding to the query feature;

[0078] The root word and the query parameter are used to obtain a corresponding second query column, and the second query column stores operational business data corresponding to the query feature.

[0079] As a preferred implementation of this embodiment, a query association formula is used to obtain the corresponding second query column, wherein the query association formula is specifically:

[0080] j=max(CV j|i +α*freq j|i )

[0081] Among them, CV j|i is the association value between the preset root i and the root j, and the association value is used to represent the degree of association between the root i and the root j, α is the query parameter, freq j|i The frequency of querying root i and root j at the same time in the preset association period is obtained by counting, CV j|i +α*freq j|i is the sum of the associated value and the frequency scaled by the query parameter, max(CV j|i +α*freq j|i ) represents taking the maximum value of the sum, j represents the root j obtained by taking the maximum value of the sum corresponding to the root i, and the second query column is obtained based on the root j.

[0082] It should be noted that the correlation value in this embodiment is a quantitative result of the correlation degree between actual word roots, which can be obtained by those skilled in the art based on well-known experience. For example, in this embodiment, CV j|i ∈[0,1] and the degree of correlation is the same as CV j|i The association period is formulated based on the actual query analysis application and can be obtained by those skilled in the art based on the known experience. For example, in this embodiment, the association period is 100 queries. For example, when querying root i for the 100th time, the administrator simultaneously queries root j for 70 times, then freq j|i =0.7, and the frequency means the data-driven association degree between root i and root j. The higher the frequency, the greater the data-driven association degree. The query parameter of this embodiment is an actual scaling value of the frequency, which can be obtained by those skilled in the art based on well-known experience. For example, in this embodiment, α∈[1,2] and the value of α is controlled by the query feedback data.

[0083] Based on the above, this embodiment provides the necessary technical means for querying through preset query rules. The correspondence between query features and word roots improves the accuracy of the query. The first query column provides corresponding data for the query to ensure the basic function of the query. The corresponding second query column is obtained through query parameters and a technical basis is provided for updating the large model. The second query column provides corresponding related data for the query to optimize the query, thereby providing a more comprehensive data basis for analysis.

[0084] Specifically, the analysis rules include a first analysis mechanism and a second analysis mechanism;

[0085] The first analysis mechanism uses the analysis feature and the first query column to perform a self-analysis corresponding to the query feature;

[0086] The second analysis mechanism uses the analysis feature, the second query column and the association parameter to perform an association analysis corresponding to the query feature; the association parameter is used to configure the analysis association degree.

[0087] It should be noted that the self-analysis in this embodiment is to select the corresponding preset analysis method based on the actual analysis needs, which can be obtained by technical personnel in this field based on well-known experience. For example, when analyzing the sales change rate, the corresponding preset sales change rate calculation formula can be used.

[0088] As a preferred implementation of this embodiment, an analysis association formula is used to determine whether to perform association analysis. The analysis association formula is specifically:

[0089]

[0090] Where M is the first query column, N is the second query column, m is the slope corresponding to the data in the first query column, n is the slope corresponding to the data in the second query column, ∑ is the summation symbol, p(m,n) represents the joint probability that the slopes of the data in the first query column and the second query column are not zero, that is, the probability that the data in the first query column and the second query column are associated, β is the association parameter, I(M,N) is the calculated association analysis index, when I(M,N)≥I τ (M,N) for correlation analysis, I τ (M, N) is the preset correlation analysis indicator threshold.

[0091] It should be noted that the association analysis index threshold of this embodiment is formulated based on the actual calculation result of the association analysis index using the existing threshold technology, which can be obtained by those skilled in the art based on the known experience. Implement data-driven construction of the association between the first query column and the second query column, and implement the The scaling of , in order to determine whether to conduct association analysis, where β is the actual The scaling value of β can be obtained by those skilled in the art based on well-known experience. For example, in this embodiment, β∈[0,1] and β is controlled by the analysis feedback data.

[0092] Based on the above, this embodiment provides the necessary technical means for analysis through preset analysis rules, realizes the basic functions of security analysis through self-analysis, realizes corresponding correlation analysis through association parameters and provides a technical basis for updating the large model, and optimizes the analysis through association analysis, thereby providing administrators with more comprehensive analysis results, which is beneficial for administrators to manage operational business data.

[0093] Specifically, the generation process of query results and query feedback includes:

[0094] A1: Obtain query features and call query rules;

[0095] A2: Use the query features obtained in step A1 to obtain the corresponding word root;

[0096] A3: Use the root word to get the corresponding first query column;

[0097] A4: Use the root word and query parameters to obtain the corresponding second query column;

[0098] A5: define the first query column obtained in step A3 and the second query column obtained in step A4 as query results;

[0099] A6: Collect the feedback content input by the administrator after receiving the query results to obtain query feedback.

[0100] In a simple example 1, for example, the query feature is "query the sales in the past three days", then the corresponding root words are "three days" and "sales", and the first query column obtained is the sales in the past three days, the CV of sales and cost 成本销售额 =0.6 and the corresponding freq 成本销售额 =0.8, at this time α=1.5, after calculation, "cost" is the root word most associated with "sales" obtained by using the query association formula, and "cost" is output, and based on "cost" the cost of the past three days recorded in the second query column is obtained.

[0101] Based on the above, the query features and query rules of this embodiment are used to obtain query results including the first query column and the second query column, thereby optimizing the query. The query feedback provides a data basis for optimizing the big model, thereby optimizing the query of operational business data based on the big model.

[0102] Specifically, the process of generating analysis results and analysis feedback includes:

[0103] B1: Obtain analysis features and call analysis rules;

[0104] B2: using the analysis feature obtained in step B1, the first analysis mechanism and the first query column to perform self-analysis corresponding to the query feature;

[0105] B3: performing association analysis corresponding to the query feature using the analysis feature obtained in step B1, the second analysis mechanism, the second query column and the association parameter;

[0106] B4: The result of the self-analysis in step B2 and the result of the association analysis in step B3 are defined as analysis results;

[0107] B5: Collect the feedback input by the administrator after receiving the analysis results to obtain analysis feedback.

[0108] Furthermore, combined with Simple Example 1, for example, based on the existence of promotional activities using discounts in the cost of the past three days, which affects the cost of the past three days, a correlation analysis is performed on the sales volume of the past three days and the cost of the past three days, and at this time β = 0.5.

[0109] Based on the above, the analysis features and analysis rules of this embodiment are used to obtain analysis results including self-analysis and association analysis, thereby optimizing the analysis. The analysis feedback provides a data basis for optimizing the big model, thereby optimizing the analysis of operational business data based on the big model.

[0110] Specifically, update the corresponding query rules and analysis rules, including:

[0111] The query parameters and the associated parameters are updated based on the query feedback and the analysis feedback.

[0112] Based on Simple Example 1, further, the feedback content input after the collection administrator receives the query results is counted. For example, in this embodiment, the frequency of the feedback content being "satisfied" is counted, and a satisfaction frequency threshold is preset. When the frequency of "satisfied" obtained by statistics is less than the satisfaction frequency threshold, it means that the current scaling value of α=1.5 is too large, then the value of α of "cost" is reduced in the next "sales" query, and adjustments are made multiple times until the frequency of "satisfied" obtained by statistics is greater than or equal to the satisfaction frequency threshold; the feedback content input after the collection administrator receives the analysis results is counted. For example, in this embodiment, the frequency of the feedback content being "satisfied" is counted, and a satisfaction frequency threshold is preset. When the frequency of "satisfied" obtained by statistics is less than the satisfaction frequency threshold, it means that the current scaling value of β=0.5 is too large, then the value of β of "cost" is reduced in the next "sales" analysis, and adjustments are made multiple times until the frequency of "satisfied" obtained by statistics is greater than or equal to the satisfaction frequency threshold.

[0113] Based on the above, this embodiment optimizes the big model by updating the query parameters and the associated parameters, thereby improving the applicability of the big model-based operational business data query and analysis method of this embodiment.

[0114] In a second aspect, this embodiment discloses Figure 2A system for querying and analyzing operational business data based on a big model is shown, and the system is applicable to the method for querying and analyzing operational business data based on a big model as described above, and the system includes:

[0115] A large model module, the large model module is configured to store and run a feature unit, a result unit, a feedback unit, and an update unit;

[0116] The feature unit is configured to: collect and parse instructions to obtain query features and analysis features;

[0117] The result unit is configured to: obtain corresponding query results and analysis results by using query features and analysis features;

[0118] The feedback unit is configured to: obtain corresponding query feedback and analysis feedback using the query result and the analysis result;

[0119] The updating unit is configured to: update corresponding query rules and analysis rules using query feedback and analysis feedback;

[0120] The big model stores preset query rules and preset analysis rules. The query rules are used to obtain query results based on the queried features, and the analysis rules are used to obtain analysis results based on the analysis features. The query features are used to characterize the query intent of the instructions, and the analysis features are used to characterize the analysis intent of the instructions. The query results and analysis results are used by administrators to manage and operate business data. The query feedback and analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and analysis results.

[0121] It should be noted that the big model-based operational business data query and analysis system of the present embodiment corresponds to the big model-based operational business data query and analysis method mentioned above. Therefore, the contents not specifically described in the big model-based operational business data query and analysis system of the present embodiment may include but are not limited to functional definitions, working principles, and technical effects, which are the same as those of the big model-based operational business data query and analysis method mentioned above, and this text will not elaborate on them.

[0122] In a third aspect, this embodiment discloses an electronic device, comprising: at least one processor and at least one memory, wherein the memory is communicatively connected to the processor; a computer program that can be executed by the processor is stored on the memory, and when the computer program is executed by the processor, the operational business data query and analysis method based on a large model as described above is implemented.

[0123] In summary, the operational business data query and analysis method, system and electronic device based on the big model of this embodiment provide technical support for operational business data query and analysis through the big model, improve the intelligence of the query analysis, provide data reference for query analysis through query features and analysis features, automatically generate query results and analysis results under the guidance of query rules and analysis rules, improve the interpretability of query analysis, and improve the applicability of the big model by updating query parameters and associated parameters, thereby realizing operational business data query and analysis based on the big model.

[0124] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein or their combination. For software implementation, part or all of the flow of the embodiment can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. Computer-readable storage media include computer storage media and communication media, wherein the communication medium includes any medium that is convenient for transmitting a computer program from one place to another. The storage medium can be any available medium that a computer can access. The computer-readable storage medium can include but is not limited to RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer.

[0125] Finally, it should be noted that the above is only a preferred embodiment of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for querying and analyzing operational business data based on a large model, characterized in that: The method includes: S1: Run a preset large model, and the large model executes steps S2 to S5; S2: Collect and parse instructions to obtain query features and analysis features; S3: using the query feature and the analysis feature to obtain corresponding query results and analysis results; S4: using the query result and the analysis result, obtaining corresponding query feedback and analysis feedback; S5: using the query feedback and the analysis feedback, updating corresponding query rules and analysis rules; The big model stores the preset query rules and the preset analysis rules, the query rules are used to obtain the query results based on the queried features, the analysis rules are used to obtain the analysis results based on the analysis features, the query features are used to characterize the query intent of the instructions, the analysis features are used to characterize the analysis intent of the instructions, the query results and the analysis results are used by the administrator to manage and operate business data, and the query feedback and the analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and the analysis results.

2. The method for querying and analyzing operational business data based on a large model according to claim 1 is characterized in that: The generation process of the query feature specifically includes: The instructions are parsed using natural language technology, and the language of the query intent related to the operational business data in the instructions is defined as the query feature.

3. The method for querying and analyzing operational business data based on a large model according to claim 1 is characterized in that: The generation process of the analysis feature specifically includes: The instructions are parsed using natural language technology, and the language of the analysis intent related to the operational business data in the instructions is defined as the analysis feature.

4. The method for querying and analyzing operational business data based on a large model according to claim 1 is characterized in that: The query rule includes a query table and query parameters; The query table includes a root, a first query column, and a second query column; The process of querying using the query table specifically includes: Using the query feature to obtain the corresponding root word; Using the root word to obtain the corresponding first query column, the first query column storing operational business data corresponding to the query feature; The corresponding second query column is obtained by using the root word and the query parameter, and the second query column stores the operational business data corresponding to the query feature.

5. The method for querying and analyzing operational business data based on a large model according to claim 4 is characterized in that: The analysis rules include a first analysis mechanism and a second analysis mechanism; The first analysis mechanism uses the analysis feature and the first query column to perform a self-analysis corresponding to the query feature; The second analysis mechanism uses the analysis feature, the second query column and an association parameter to perform an association analysis corresponding to the query feature; the association parameter is used to configure the analysis association degree.

6. The method for querying and analyzing operational business data based on a large model according to claim 4 is characterized in that: The generation process of the query result and the query feedback specifically includes: A1: Acquire the query feature and call the query rule; A2: using the query feature obtained in step A1 to obtain the corresponding word root; A3: Using the root word to obtain the corresponding first query column; A4: Obtain the corresponding second query column using the root word and the query parameter; A5: define the first query column obtained in step A3 and the second query column obtained in step A4 as the query result; A6: Collect the feedback content input by the administrator after receiving the query result to obtain the query feedback.

7. The method for querying and analyzing operational business data based on a large model according to claim 5 is characterized in that: The generation process of the analysis result and the analysis feedback specifically includes: B1: Acquire the analysis feature and call the analysis rule; B2: using the analysis feature obtained in step B1, the first analysis mechanism and the first query column to perform the self-analysis corresponding to the query feature; B3: performing association analysis corresponding to the query feature using the analysis feature obtained in step B1, the second analysis mechanism, the second query column and the association parameter; B4: defining the result of the self-analysis in step B2 and the result of the association analysis in step B3 as the analysis result; B5: Collecting feedback content input by the administrator after receiving the analysis result to obtain the analysis feedback.

8. The method for querying and analyzing operational business data based on a large model according to claim 6 or 7, characterized in that: The query rules and analysis rules corresponding to the update specifically include: The query parameters and the association parameters are updated based on the query feedback and the analysis feedback.

9. A system for querying and analyzing operational business data based on a big model, the system being applicable to the method for querying and analyzing operational business data based on a big model as claimed in any one of claims 1 to 8, characterized in that: The system includes: A large model module, wherein the large model module is configured to store and run a feature unit, a result unit, a feedback unit, and an update unit; The feature unit is configured to: collect and parse instructions to obtain query features and analysis features; The result unit is configured to: obtain corresponding query results and analysis results using the query characteristics and the analysis characteristics; The feedback unit is configured to: obtain corresponding query feedback and analysis feedback using the query result and the analysis result; The updating unit is configured to: update corresponding query rules and analysis rules using the query feedback and the analysis feedback; The big model stores the preset query rules and the preset analysis rules, the query rules are used to obtain the query results based on the queried features, the analysis rules are used to obtain the analysis results based on the analysis features, the query features are used to characterize the query intent of the instructions, the analysis features are used to characterize the analysis intent of the instructions, the query results and the analysis results are used by the administrator to manage and operate business data, and the query feedback and the analysis feedback are used to characterize the administrator's satisfaction with the accuracy of the query results and the analysis results.

10. An electronic device, characterized in that: include: At least one processor and at least one memory, the memory being communicatively connected to the processor; the memory storing a computer program that can be executed by the processor, and when the computer program is executed by the processor, the large model-based operational business data query and analysis method as described in any one of claims 1 to 8 is implemented.

Images