A sales data statistics method, system and storage medium based on AIGC

Abstract

The present application relates to the field of sales data statistics, and in particular to a sales data statistics method, system and storage medium based on AIGC, which include reading wide table setting data, determining wide table data through the wide table setting data; reading query data, determining required wide table data through the query data and preset AIGC, and determining an SQL query statement through the query data, the AIGC and the required wide table data; determining the grammatical correctness of the SQL query statement through a preset compiler and the SQL query statement, and determining the semantic correctness of the SQL query statement through the AIGC and the SQL query statement; determining result data through the SQL query statement and the required wide table data; determining format data through the AIGC and the result data, and determining output data through the format data and the result data. The present application has the effect of lowering the threshold for user use.

Description

Technical Field

[0001] The present application relates to the field of sales data statistics, and in particular to a sales data statistics method, system and storage medium based on AIGC. Background Art

[0002] In order to manage sales and improve efficiency, enterprises will use order systems and accumulate sales data in the order systems. When enterprise employees or management need to query and count sales data, they often need to log in to the corresponding order system, find the corresponding function, enter the query conditions, and query.

[0003] Data warehouses and BI systems are commonly used nowadays. First, the data in each system is extracted and split into multiple data dimensions and stored in the data warehouse. Then, the data in the data warehouse is read and called through the BI system, and statistical reports are generated through the visual report component. However, if the user needs a new data dimension, and this data dimension does not exist in the current data warehouse, if the user does not master the relevant technology, the BI system cannot be used to perform secondary processing on the original data to generate the required data. Professional technicians are required to use the BI system to develop data, or add data dimensions to the data warehouse, and then provide them to users. Therefore, users are required to have professional capabilities, but general corporate employees (such as sales managers and business personnel) do not have professional IT capabilities, and data warehouses require a lot of development costs and later maintenance costs, resulting in high IT costs. Summary of the invention

[0004] In order to improve the problem of user data warehouse, the present application provides a sales data statistics method, system and storage medium based on AIGC.

[0005] The present application provides a sales data statistics method and system based on AIGC, which adopts the following technical solutions:

[0006] A sales data statistical method based on AIGC, comprising:

[0007] Reading wide table setting data, and determining wide table data according to the wide table setting data;

[0008] Read query data, determine required wide table data through the query data and a preset AIGC, and determine an SQL query statement through the query data, the AIGC, and the required wide table data;

[0009] Determine the syntax correctness of the SQL query statement through a preset compiler and the SQL query statement, and determine the semantic correctness of the SQL query statement through the AIGC and the SQL query statement;

[0010] Determine result data through the SQL query statement and the required wide table data;

[0011] Format data is determined by the AIGC and the result data, and output data is determined by the format data and the result data.

[0012] By adopting the above technical solutions, the design of wide tables is much simpler than the original data warehouse design. There is no need to disassemble data dimensions for secondary data processing, which reduces the workload. There is no need to design the relationship between data subjects and dimensions, which greatly reduces the complexity of data structures and storage space requirements. Wide tables are directly mapped to the original data of each business system, so technical engineers can intuitively understand data relationships, simplifying development and operation and maintenance, greatly reducing the technical threshold required for users to use, and reducing sunk learning costs. Users only need to issue spoken instructions without any system operations, which greatly improves the operating experience. Data statistics and formatted output are completed based on AIGC, which does not require the work of any technical personnel and can meet any personalized data statistics needs of users.

[0013] Optionally, after determining the output data through the format data and the result data, the method further includes:

[0014] Feedback data is read, and the AIGC is adjusted according to the feedback data.

[0015] By adopting the above technical solution, the user inputs feedback data after receiving the output data representing the query result, and then AIGC is trained accordingly through the feedback data, so that subsequent query results are more accurate, which plays a feedback role.

[0016] Optionally, adjusting the AIGC according to the feedback data includes:

[0017] If the feedback data read is unsatisfactory, the result data and the query data corresponding to the result data are stored in a preset negative case library. After the query data is read and the result data is generated, the same query data is retrieved from the negative case library. If the query data retrieved corresponds to the result data, the AIGC is fine-tuned within the range of the preset floating data to obtain different result data.

[0018] By adopting the above technical solution, erroneous result data is recorded through a negative case library, and the output of erroneous result data is reduced by retrieving the negative case library, thereby improving accuracy.

[0019] Optionally, adjusting the AIGC according to the feedback data further includes:

[0020] A correction number data is preset, and the number of times the query data is read is counted. If the number of times the query data is read falls within the correction number data range, the result data corresponding to the query data read this time is retrieved from the negative case library and directly output. If the read feedback data is satisfactory, the result data and the query data corresponding to the result data are removed from the negative case library; if the read feedback data is unsatisfactory, the result data and the query data corresponding to the result data are moved to a preset error database.

[0021] By adopting the above technical solution, the result data in the negative case library can be reproduced by correcting the frequency data, thereby reducing the probability that the feedback data maliciously input by individual users interferes with the training of AIGC and improving accuracy; the misjudged result data is removed from the negative case library, and the truly problematic result data is further moved into the error database, which also reduces the probability that the truly problematic result data will be reproduced during reproduction, resulting in a decrease in user usage.

[0022] Optional, preset error database, including:

[0023] Preset error count data, determine database trigger data through the correction count data and the error count data, there are multiple error databases, the error databases are arranged in order, each error database corresponds to one database trigger data, and the next database trigger data is determined through the previous database trigger data and the error count data. When the number of times the query data is read falls within the database trigger data range, read the result data from the corresponding error database. If the read feedback data is satisfactory, move the result data and the query data corresponding to the result data to the previous error database. If they were originally in the first error database, move them to the negative case library. If the read feedback data is unsatisfactory, move the result data and the query data corresponding to the result data to the next error database.

[0024] By adopting the above technical solution, the probability of correct result data entering the error database due to multiple misjudgments is reduced, and the correct result data in the error database is removed through reproduction, thereby further improving accuracy and reducing the influence of the user's personal initiative on the objective results.

[0025] Optionally, a preset error database also includes:

[0026] When the result data corresponding to the query data in the negative case library is read and the feedback data read is unsatisfactory, the database trigger data is determined by the correction number data and the error number data, and the database trigger data is used to retrieve whether there is a corresponding error database. If so, the query data and the corresponding result data are moved into the error database. If not, the error database is established, and the database trigger data is assigned to the established error database, and then the query data and the corresponding result data are moved into the error database.

[0027] By adopting the above technical solution, an error database is automatically established, which reduces the user's workload and further lowers the technical threshold required by the user.

[0028] Optionally, a preset error database also includes:

[0029] When the result data corresponding to the query data in the error database is read and the feedback data read is unsatisfactory, new database trigger data is determined by the database trigger data corresponding to the error database read and the error count data, and the new database trigger data is used to retrieve whether there is a corresponding error database. If so, the query data and the corresponding result data are moved to the next error database; if not, a new error database is established, and the new database trigger data is assigned to the newly established error database, and then the query data and the corresponding result data are moved to the next error database.

[0030] By adopting the above technical solution, multiple error databases are automatically established, and the probability of recurrence of result data that have been judged to be erroneous multiple times is reduced, the accuracy is improved, and the probability of reduced user experience caused by reading result data in the error database is reduced.

[0031] This application provides a sales data statistics system based on AIGC, which adopts the following technical solutions:

[0032] A sales data statistics system based on AIGC, comprising:

[0033] The operation panel is used for human-computer interaction with the user, for the user to input wide table setting data, query data, and form and output operation signals;

[0034] The processor receives the operation signal, performs calculations on the signal and data, and outputs the result data;

[0035] A database, used to store data and signals and to provide access to the processor;

[0036] The display module receives the result data and displays it to the user.

[0037] By adopting the above technical solution, human-computer interaction between the user and the processor is realized through the operation panel, data is stored in the database and called by the processor for calculation, and the results are output to the user for viewing through the display module, which is convenient and fast.

[0038] The present application provides a computer-readable storage medium, which adopts the following technical solution:

[0039] A computer-readable storage medium stores a computer program of a sales data statistics method based on AIGC that can be loaded and executed by a processor.

[0040] By adopting the above technical solution, the computer program is stored in a computer-readable storage medium.

[0041] In summary, the present application includes at least one of the following beneficial technical effects:

[0042] 1. The wide table design reduces the complexity and space requirements of data storage, and reduces the workload of data development and operation and maintenance.

[0043] 2. Users only need to issue spoken commands without any system operations, which greatly improves the operating experience.

[0044] 3. Complete data statistics and formatted output based on AIGC, without the need for any technical personnel, to meet any personalized data statistics needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of a sales data statistics method based on AIGC in an embodiment of the present application.

[0046] Figure 2 It is a flowchart diagram after step S14.

[0047] Figure 3 It is a flowchart diagram of step S5.

[0048] Figure 4 It is a module diagram of a system for sales data statistics based on AIGC in an embodiment of the present application.

[0049] Explanation of the accompanying drawings: 1. operation panel; 2. processor; 3. database; 4. display module. DETAILED DESCRIPTION

[0050] The following is combined with Figure 1-4 This application is described in further detail.

[0051] The present application embodiment discloses a sales data statistics method and system based on AIGC. Figure 1 , the sales data statistical method based on AIGC includes the following steps:

[0052] S1. The user inputs wide table setting data, reads the wide table setting data, and determines the wide table data through the wide table setting data;

[0053] S11, the user inputs query data, reads the query data, determines the required wide table data through the query data and the preset AIGC, and determines the SQL query statement through the query data, AIGC and the required wide table data;

[0054] S12, determining the grammatical correctness of the SQL query statement through a preset compiler and the SQL query statement, and determining the semantic correctness of the SQL query statement through AIGC and the SQL query statement;

[0055] S13, determining result data through SQL query statements and required wide table data;

[0056] S14. Determine format data through AIGC and result data, and determine output data through the format data and result data.

[0057] For example, the user inputs wide table setting data to set the wide table. For example, a wide table can be formulated for each subject, and all cross-system data related to the subject can be stored in the wide table. For example, the order subject wide table contains order detail data in the wholesale order system, customer details in the CRM customer relationship management system, and delivery detail data in the TMS logistics system. In this case, the input wide table setting data contains at least three data (a, b, c), where a is the subsystem data, b is the data type, and c is the data quantity, for example (wholesale order system, ignition device, 1000);

[0058] The user inputs query data for querying, for example, inputs "I want to see the sales situation in Shandong Province this year". In this embodiment, AIGC is trained and optimized for the cycle of continuously reading query data and obtaining result data. AIGC analyzes the query data input by the user, and determines which corresponding wide table data (one or more) are needed in combination with all wide table data, that is, the required wide table data, thereby obtaining an SQL query statement, and then the compiler and AIGC perform a credibility query on the correctness of the syntax and semantics of the SQL query statement; if there is a problem with the credibility query, this is output to AIGC as an error generation result, so that AIGC performs training optimization and generates SQL query statements again through AIGC until the credibility query is correct; then the SQL query statement is executed in the previously determined required wide table data (that is, one or more corresponding wide table data), thereby obtaining result data, and then outputs it to the user for display according to the format data, for example, outputs the corresponding bar chart, pie chart and other different forms for display to the user.

[0059] Reference Figure 2 , after step S14, the following steps are also included:

[0060] S2. Read feedback data and adjust AIGC based on the feedback data;

[0061] S21. If the feedback data read is unsatisfactory, the result data and the query data corresponding to the result data are stored in a preset negative case library. After the query data is read and the result data is generated, the same query data is retrieved from the negative case library. If the result data corresponding to the retrieved query data is satisfactory, the AIGC is fine-tuned within the range of the preset floating data to obtain different result data.

[0062] For example: If the query data input by the user is "I want to see the sales situation in Shandong Province this year", the result is a bar chart representing the sales situation in Shaanxi Province. The user enters dissatisfied feedback data. At this time, the query data is "I want to see the sales situation in Shandong Province this year", and the result data is "Sales situation in Shaanxi Province". The two data are stored in the negative case library. When subsequent users enter similar query data, such as "I want to see the sales situation in Shandong Province this year", "How is the sales situation in Shandong Province this year", etc., if the generated result data is still "Sales situation in Shaanxi Province", that is, the result data generated this time is the same as the result data in the negative case library, the algorithm parameters of AIGC are automatically fine-tuned to generate different result data. If the result data is different from the result data in the negative case library, it is output to the user.

[0063] Reference Figure 2 , after step S21, the following steps are also included:

[0064] S3. Preset the number of corrections and count the number of times the query data is read. If the number of times the query data is read falls within the range of the number of corrections, the result data corresponding to the query data read this time is retrieved from the negative case library and directly output. If the feedback data read is satisfactory, the result data and the query data corresponding to the result data are removed from the negative case library; if the feedback data read is unsatisfactory, the result data and the query data corresponding to the result data are moved to the preset error database.

[0065] For example, if the correction times data is set to 10,000, it means that after reading the query data for every 10,000 times, if the query data input for the next time exists in the negative case library, the result data will be read from the negative case library. If the query data input for the next time does not find a consistent result in the negative case library, the query will continue to be read until the next time, until a consistent result is found in the negative case library.

[0066] That is, after the user has input query data 10,000 times, the next query data input is "I want to see the sales situation in Shandong Province this year". If the same query data that has been collected is not retrieved from the negative case library, the result data is generated normally; if the same or similar query data is retrieved from the negative case library, such as "I want to see the sales situation in Shandong Province this year", the corresponding result data is directly read from the negative case library. If multiple corresponding result data are collected, one is read randomly. At this time, if the feedback data entered by the user is read as satisfactory, it means that the user may have made a misjudgment before, and the result data and query data are removed from the negative case library; if the feedback data entered by the user is still unsatisfactory, it further confirms that there is indeed a problem with the result data, and the result data and query data are moved to the error database.

[0067] Reference Figure 2 , after step S3, the following steps are also included:

[0068] S4. Preset error count data, determine database trigger data through correction count data and error count data, there are multiple error databases, the error databases are arranged in order, each error database corresponds to one database trigger data, determine the next database trigger data through the previous database trigger data and error count data, when the number of times the query data is read falls within the database trigger data range, read the result data from the corresponding error database, if the read feedback data is satisfactory, move the result data and the query data corresponding to the result data to the previous error database, if it was originally in the first error database, move it to the negative case library; if the read feedback data is unsatisfactory, move the result data and the query data corresponding to the result data to the next error database.

[0069] For example: if the number of errors is 12345, the database trigger data is calculated as 10000+12345=22345 by combining the correction number and the error number. In this embodiment, addition is used. Alternatively, the number of errors can be set to 2 and calculated with the correction number using multiplication or sub-methods. In this embodiment, addition is used for convenience of example calculation. The database trigger data indicates that the query data is read every 22345 times, and the next query data will read the result data from the corresponding error database. There are multiple error databases, which are named as error database No. 1, error database No. 2, error database No. 3, etc. in this embodiment and are arranged in order. The database trigger data of 22345 is assigned to error database No. 1, and then the database trigger data of error database No. 1 is added. The new database trigger data is calculated with the error number data as 22345+12345=34690. The database trigger data 34690 is assigned to the error database No. 2, and the database trigger data 47035 is assigned to the error database No. 3. The error number data is set to use a multiple of the non-correction number data or to have as few identical factors as possible to avoid the subsequent database trigger data being the same as the correction number data. If the database trigger data is the same as the correction number data, the database trigger data has a higher priority, and when the query data read exceeds the maximum database trigger data, it will reset and start counting from the beginning, that is, when the query data reaches 47035, after the result data is read from the error database No. 3, the query data count will start counting again from 0;

[0070] If the number of times the query data is read is 34690, the result data will be read from the error database No. 2 when the query data is read next time. If the feedback data of the user input read this time is satisfactory, the result data and the query data will be moved from the error database No. 2 to the error database No. 1; if the feedback data of the user input read this time is unsatisfactory, the result data and the query data will be moved from the error database No. 2 to the error database No. 3;

[0071] If there is no error database in a later order, that is, the number of times the data is queried is 47035, and the feedback data read is unsatisfactory, but there is no error database No. 4, the result data and the query data will not be moved.

[0072] Reference Figure 3 , after step 4, the following steps are also included:

[0073] S5. When the result data corresponding to the query data in the negative case library is read and the feedback data read is unsatisfactory, the database trigger data is determined by the correction number data and the error number data, and the database trigger data is used to retrieve whether there is a corresponding error database. If there is, the query data and the corresponding result data are moved into the error database. If not, an error database is established, and the database trigger data is assigned to the established error database, and then the query data and the corresponding result data are moved into the error database;

[0074] S51. When the result data corresponding to the query data in the error database is read and the feedback data read is unsatisfactory, new database trigger data is determined by the database trigger data corresponding to the read error database and the error count data, and the new database trigger data is used to retrieve whether there is a corresponding error database. If so, the query data and the corresponding result data are moved to the next error database; if not, a new error database is established, and the new database trigger data is assigned to the newly established error database, and then the query data and the corresponding result data are moved to the next error database.

[0075] For example: the user does not need to actively establish an error database. Instead, after reading the result data from the negative case library, if the feedback data read is unsatisfactory, the correction number data and the error number data are calculated to obtain the error database in which the result data should be stored. The corresponding database trigger data is 10000+12345=22345. However, if the error database does not exist, an error database can be automatically established through AIGC or other methods, and the database trigger data 22345 is assigned to the newly established error database, and then the result data and query data are moved to this error database; if the result data is read from the error database No. 1 and the feedback data read is unsatisfactory, the database trigger data The database trigger data corresponding to the error database in which the result data calculated from the error number data should be stored is 22345+12345=34690. However, when this error database does not exist, an error database can be automatically established through AIGC or other means, and the database trigger data of 34690 is assigned to the newly established error database (error database No. 2). The result data and query data of this time are then moved into this error database. The user initially sets the quantity threshold data of the error database. If the quantity threshold data is set to 3, the error database will be established up to error database No. 3 at most. Even if error database No. 4 does not exist, no new error database will be established to limit the number of error databases.

[0076] The implementation principle of a sales data statistics method based on AIGC in an embodiment of the present application is: establish wide table data by inputting wide table setting data, then input query data in a spoken language, determine professional SQL query statements through AIGC, then determine the grammatical and semantic correctness of the SQL query statements through the compiler and AIGC, then obtain result data through the SQL query statements and wide table data, and determine the output data to the user according to the format data.

[0077] A sales data statistics system based on AIGC, referring to Figure 4 , including an operation panel 1, a processor 2, a database 3 and a display module 4. In this embodiment, the operation panel 1 can be a key, or a keyboard including a plurality of keys, or a touch screen and other devices for users to input data for human-computer interaction. The display module 4 can be a digital screen, a display screen or a touch screen and other devices that can perform optical display. The user inputs wide table setting data, query data, error number data, correction number data and other data and signals through the operation panel 1, and forms an operation signal to output to the processor 2. The processor 2 calls relevant data and signals from the database 3, and obtains result data after calculating and processing the operation signal. The result data obtains output data according to the format data and is displayed through the display module 4.

[0078] Processor 2 may include a central processing unit such as a CPU or MPU or a host system built around a CPU or MPU, including hardware or software. After the meter has a processor 2, people can freely control the metering instrument by programming so that it can run according to people's wishes. Processor 2 can control local measurement transmission, remote measurement transmission, remote communication components, etc. through internal protocols. Internal protocols generally refer to all protocols that achieve mutual communication or links within the same metering instrument or the same system, including: human-computer interaction protocols, software / hardware (interface) protocols, chip bus (C-Bus) protocols, internal bus (I-Bus) protocols, etc., some or all of the protocols. With the development of integrated circuit technology, some protocols that belong to the external bus (E-Bus) protocol are also classified as internal protocols after the external bus (E-Bus) is integrated into the chip.

[0079] A computer-readable storage medium stores a computer program of a sales data statistics method based on AIGC that can be loaded and executed by a processor.

[0080] Computer-readable storage media include, for example, various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0081] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.

Claims

1. A sales data statistics method based on AIGC, characterized in that: include: Reading wide table setting data, and determining wide table data according to the wide table setting data; Read query data, determine required wide table data through the query data and a preset AIGC, and determine an SQL query statement through the query data, the AIGC, and the required wide table data; Determine the syntax correctness of the SQL query statement through a preset compiler and the SQL query statement, and determine the semantic correctness of the SQL query statement through the AIGC and the SQL query statement; Determine result data through the SQL query statement and the required wide table data; Determine format data through the AIGC and the result data, and determine output data through the format data and the result data; Reading feedback data, and adjusting the AIGC according to the feedback data; If the feedback data read is unsatisfactory, the result data and the query data corresponding to the result data are stored in a preset negative case library, and after the query data is read and the result data is generated, the same query data is retrieved from the negative case library, and if the result data corresponding to the query data is retrieved, the AIGC is fine-tuned within the range of the preset floating data to obtain different result data; Preset correction number data, and count the number of times the query data is read. If the number of times the query data is read falls within the correction number data range, the result data corresponding to the query data read this time is retrieved from the negative case library and directly output. If the feedback data read is satisfactory, the result data and the query data corresponding to the result data are removed from the negative case library; If the feedback data read is unsatisfactory, the result data and the query data corresponding to the result data are moved to a preset error database; Preset error database, including: Preset error count data, determine database trigger data through the correction count data and the error count data, there are multiple error databases, the error databases are arranged in order, each error database corresponds to one database trigger data, and the next database trigger data is determined through the previous database trigger data and the error count data. When the number of times the query data is read falls within the database trigger data range, read the result data from the corresponding error database. If the read feedback data is satisfactory, move the result data and the query data corresponding to the result data to the previous error database. If they were originally in the first error database, move them to the negative case library. If the read feedback data is unsatisfactory, move the result data and the query data corresponding to the result data to the next error database.

2. The sales data statistics method based on AIGC according to claim 1, characterized in that: The default error database also includes: When the result data corresponding to the query data in the negative case library is read and the feedback data read is unsatisfactory, the database trigger data is determined by the correction number data and the error number data, and the database trigger data is used to retrieve whether there is a corresponding error database. If so, the query data and the corresponding result data are moved into the error database. If not, the error database is established, and the database trigger data is assigned to the established error database, and then the query data and the corresponding result data are moved into the error database.

3. The sales data statistics method based on AIGC according to claim 2 is characterized in that: The default error database also includes: When the result data corresponding to the query data in the error database is read and the feedback data read is unsatisfactory, new database trigger data is determined by the database trigger data corresponding to the error database read and the error count data, and the new database trigger data is used to retrieve whether there is a corresponding error database. If so, the query data and the corresponding result data are moved to the next error database; if not, a new error database is established, and the new database trigger data is assigned to the newly established error database, and then the query data and the corresponding result data are moved to the next error database.

4. A system using the AIGC-based sales data statistics method according to any one of claims 1 to 3, characterized in that: include: An operation panel (1) is used for human-computer interaction with a user, for allowing the user to input wide table setting data, query data, and generate and output operation signals; The processor (2) receives the operation signal, performs calculation processing on the signal and data, obtains result data, and determines output data according to the format data; A database (3) for storing data and signals and providing them for the processor (2) to call; The display module (4) receives the output data and displays it to the user.

5. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and executes a sales data statistics method based on AIGC as described in any one of claims 1 to 3.

Images