Method and device for processing data in table, electronic equipment and storage medium

Abstract

The invention relates to a method and a device for processing data in a table, electronic equipment and a storage medium. The method comprises the following steps: acquiring original data in the table; obtaining an initial processing result corresponding to the original data through an initial artificial intelligence model; in response to a modification operation for the initial processing result, obtaining the modified initial processing result; wherein the modified initial processing result is used for being provided for the initial artificial intelligence model to learn the incidence relation between the original data and the modified initial processing result, so that a target artificial intelligence model is obtained. Therefore, the accuracy of the processing result output by the artificial intelligence model can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, electronic device and storage medium for processing data in a table. Background Art

[0002] At present, users, enterprises, etc. generate a large amount of tabular data in their operations and need to process them in batches. Due to the complexity of tabular data, existing processing methods often have certain limitations, such as using functions but making it difficult to write rules. For this reason, an artificial intelligence model is introduced to process a large amount of tabular data in batches.

[0003] In the related art, the artificial intelligence model needs to be pre-trained with table data in advance, and then the artificial intelligence model can be used to batch process the user's table data. However, since the user's table data and the table data involved in the artificial intelligence model pre-training cannot completely match, the processing results output by the artificial intelligence model will be inaccurate. Summary of the invention

[0004] In order to solve the above technical problem that the processing results output by the artificial intelligence model are inaccurate because the user's table data and the table data participating in the pre-training of the artificial intelligence model cannot be completely matched, the embodiment of the present application provides a method, device, electronic device and storage medium for processing data in a table. The specific technical solution is as follows:

[0005] In a first aspect of an embodiment of the present application, a sample collection method is first provided, the method comprising:

[0006] Obtain the original data in the table; obtain the initial processing results corresponding to the original data through the initial artificial intelligence model; in response to the modification operation on the initial processing result, obtain the modified initial processing result; wherein the modified initial processing result is used to provide the initial artificial intelligence model with learning the association between the original data and the modified initial processing result, so as to obtain the target artificial intelligence model.

[0007] In an optional embodiment, the obtaining of the original data in the table includes: obtaining the table, and displaying an artificial intelligence setting page in response to a confirmation operation of artificial intelligence processing; wherein the artificial intelligence setting page is used to set the data source, the processing result filling position and the label; in response to the setting operation of the data source, determining the set original data in the table; in response to the setting operation of the label, determining the set label; in response to the setting operation of the processing result filling position, determining the set processing result filling position; in response to the processing result preset operation, obtaining the original data in the table; the obtaining of the initial processing result corresponding to the original data through the initial artificial intelligence model includes: using the initial artificial intelligence model to process the original data according to the label to obtain the corresponding initial processing result.

[0008] In an optional embodiment, in response to a preset operation of processing results, obtaining the original data in the table includes: in response to a processing result preview operation, obtaining the original data in the table; in response to a modification operation on the initial processing result, obtaining the modified initial processing result includes: displaying the original data in the table and the initial processing result corresponding to the original data on a preview result page; in response to a modification operation on the initial processing result, obtaining the modified initial processing result.

[0009] In an optional embodiment, the obtaining of the original data in the table in response to a processing result preset operation includes: obtaining the original data in the table in response to a processing result generation operation; the obtaining of the modified initial processing result in response to a modification operation on the initial processing result includes: filling the initial processing result corresponding to the original data in the processing result fill position; obtaining the modified initial processing result in response to a modification operation on the initial processing result.

[0010] In an optional embodiment, in response to a modification operation on the initial processing result, after obtaining the modified initial processing result, the method also includes: displaying a first prompt information, the first prompt information being used to prompt to provide a processing sample to the initial artificial intelligence model; in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction; in response to the first sample collection instruction, generating a first processing sample based on the original data and the modified initial processing result; and training the initial artificial intelligence model based on the first processing sample to obtain a target artificial intelligence model.

[0011] In an optional embodiment, in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model and triggering a sample collection instruction, the method further includes: in response to a second sample collection instruction, displaying a processing sample configuration page, the processing sample configuration page being used to configure a second processing sample; on the processing sample configuration page, obtaining the target original data in the table input by the user, and the target processing result corresponding to the target original data; generating a second processing sample based on the target original data and the target processing result corresponding to the target original data; and training the initial artificial intelligence model based on the second processing sample to obtain a target artificial intelligence model.

[0012] In an optional embodiment, the generating of the first processing sample according to the original data and the modified initial processing result includes: determining the modified initial processing result as the first expected processing result corresponding to the original data; generating a forward processing sample according to the original data and the first expected processing result corresponding to the original data. In an optional embodiment, the generating of the first processing sample according to the original data and the modified initial processing result includes: determining the initial processing result before modification corresponding to the modified initial processing result; combining the preset reverse prompt information with the initial processing result before modification to generate a second expected processing result; generating a reverse processing sample according to the original data and the second expected processing result.

[0013] In an optional embodiment, the artificial intelligence settings page is displayed in response to the confirmation operation of the artificial intelligence processing, including: in response to the selection operation of a cell in a table, a second prompt message is displayed, wherein the second prompt message is used to prompt that artificial intelligence will be used for processing, and there are multiple pairs of duplicate labels in the cells in the column where the cell is located; in response to the confirmation operation of the artificial intelligence processing, the artificial intelligence settings page is displayed.

[0014] In an optional embodiment, before displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page, the method also includes: displaying the target original data in the table and the target processing results corresponding to the target original data on the preview result page; wherein the target original data is the original data in the table processed by the user, and the target processing results are the multiple pairs of repeated labels; in the preview result page, displaying a third prompt message, the third prompt message is used to prompt to provide a processing sample to the initial artificial intelligence model; in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction; in response to the third sample collection instruction, acquiring the target original data and the target processing result corresponding to the target original data; generating a third processing sample based on the target original data and the target processing result corresponding to the target original data; and training the initial artificial intelligence model based on the third processing sample to obtain a target artificial intelligence model.

[0015] In an optional embodiment, the method further includes: in response to a processing result generation operation, obtaining the original data in the table; processing the original data according to the label through the target artificial intelligence model to obtain a corresponding new processing result; and filling the new processing result corresponding to the original data in the processing result fill position.

[0016] In an optional embodiment, the method also includes: in response to an instruction creation operation, displaying an instruction creation page, wherein the instruction creation page is used to set the persona, examples and tasks in the instruction; in response to the persona setting operation, determining the set persona, wherein the persona is used to limit the working scope of the initial artificial intelligence model; in response to the example setting operation, determining the set example, wherein the example is used to instruct the initial artificial intelligence model to perform data processing with reference to the example; in response to the task setting operation, determining the set task, wherein the task is used to instruct the initial artificial intelligence model to process the target original data in the table; in response to the processing result generation operation, providing the persona, examples and tasks in the instruction to the initial artificial intelligence model; and filling the processing result fill position of the table with the processing result of the initial artificial intelligence model on the target original data in the table.

[0017] In an optional embodiment, the display of the instruction creation page in response to the instruction creation operation includes: displaying the artificial intelligence settings page in response to the artificial intelligence processing selection operation, wherein the artificial intelligence settings page is used to set the data source, the processing result filling position and the label; in response to the instruction center selection operation, the instruction center page is displayed on the artificial intelligence settings page; in response to the instruction creation operation, the instruction creation page is displayed on the instruction center page; in response to the processing result generation operation, the persona, examples and tasks in the instruction are provided to the initial artificial intelligence model, including: displaying the instruction center page in response to the instruction confirmation operation; on the instruction center page, in response to the confirmation operation, the artificial intelligence settings page is displayed; in response to the processing result generation operation, the persona, examples and tasks in the instruction are provided to the initial artificial intelligence model.

[0018] In an optional embodiment, the displaying of the instruction creation page in response to the instruction creation operation includes: in response to the processing result preview operation, displaying the original data in the table and the initial processing result corresponding to the original data on the preview result page; in response to the instruction viewing operation, displaying the instruction details page on the preview result page; in response to the instruction creation operation, displaying the instruction creation page on the instruction details page; in response to the processing result generation operation, providing the persona, examples and tasks in the instruction to the large language model, including: in response to the instruction confirmation operation, displaying the instruction center page; in response to the confirmation operation on the instruction center page, displaying the preview result page; in response to the page jump operation, jumping from the preview result page to the artificial intelligence setting page; in the artificial intelligence setting page, in response to the processing result generation operation, providing the persona, examples and tasks in the instruction to the initial artificial intelligence model.

[0019] In an optional embodiment, the example includes tabular data, a label, and an output position, and determining the set example in response to an example setting operation includes: determining the example original data in the set table in response to the tabular data setting operation; determining the label corresponding to the set example original data in response to the label setting operation; and determining the fill position corresponding to the processing result of the set example original data in response to the output position setting operation.

[0020] In an optional embodiment, the task includes tabular data, a label, and an output position, and the determination of the set task in response to the task setting operation includes: determining the target original data in the set table in response to the tabular data setting operation; determining the label corresponding to the set target original data in response to the label setting operation; and determining the filling position of the processing result of the set target original data in response to the output position setting operation.

[0021] In a second aspect of the embodiments of the present application, a data processing device in a table is further provided, the device comprising:

[0022] A data acquisition module is used to obtain the original data in the table;

[0023] A data processing module, used to obtain an initial processing result corresponding to the original data through an initial artificial intelligence model;

[0024] A result modification module, used for obtaining a modified initial processing result in response to a modification operation on the initial processing result;

[0025] Among them, the modified initial processing result is used to provide the initial artificial intelligence model with a learning relationship between the original data and the modified initial processing result to obtain a target artificial intelligence model.

[0026] In an optional implementation, the data acquisition module specifically includes:

[0027] a page display submodule, for acquiring a table and, in response to a confirmation operation of the artificial intelligence processing, displaying an artificial intelligence setting page;

[0028] The artificial intelligence setting page is used to set the data source, the processing result filling position and the label;

[0029] The original data setting submodule is used to determine the original data in the set table in response to the setting operation of the data source;

[0030] The label setting submodule is used to determine the set label in response to the label setting operation;

[0031] A filling position setting submodule, for determining a set processing result filling position in response to a setting operation of the processing result filling position;

[0032] A data acquisition submodule, used for acquiring the original data in the table in response to a preset operation of the processing result;

[0033] The data processing module is specifically used for:

[0034] The initial artificial intelligence model is used to process the raw data according to the labels to obtain corresponding initial processing results.

[0035] In an optional implementation, the data acquisition submodule is specifically used for:

[0036] In response to the processing result preview operation, obtaining the original data in the table;

[0037] The result modification module is specifically used for:

[0038] Displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page;

[0039] In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

[0040] In an optional implementation, the data acquisition submodule is specifically used for:

[0041] In response to the processing result generating operation, obtaining the original data in the table;

[0042] The result modification module is specifically used for:

[0043] Fill the initial processing result corresponding to the original data in the processing result filling position;

[0044] In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

[0045] In an optional embodiment, the device further comprises:

[0046] A first prompt information display module, used to display first prompt information, wherein the first prompt information is used to prompt the initial artificial intelligence model to provide a processing sample;

[0047] An instruction triggering module, configured to trigger a sample collection instruction in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model;

[0048] A first processed sample generating module, configured to generate a first processed sample according to the original data and the modified initial processing result in response to a first sample collecting instruction;

[0049] The first model training module is used to train the initial artificial intelligence model based on the first processed sample to obtain a target artificial intelligence model.

[0050] In an optional embodiment, the device further comprises:

[0051] a configuration page display module, configured to display a processing sample configuration page in response to the second sample collection instruction, wherein the processing sample configuration page is used to configure the second processing sample;

[0052] A data and result acquisition module, used to acquire the target original data in the table input by the user and the target processing result corresponding to the target original data on the processing sample configuration page;

[0053] A second processing sample generating module, configured to generate a second processing sample according to the target original data and a target processing result corresponding to the target original data;

[0054] The second model training module is used to train the initial artificial intelligence model based on the second processed sample to obtain a target artificial intelligence model.

[0055] In an optional implementation, the first processed sample generating module is specifically configured to:

[0056] Determining the modified initial processing result as the first expected processing result corresponding to the original data;

[0057] A forward processing sample is generated according to the original data and the first expected processing result corresponding to the original data.

[0058] In an optional implementation, the first processed sample generating module is specifically configured to:

[0059] Determine the initial processing result before modification corresponding to the initial processing result after modification;

[0060] Combining the preset reverse prompt information with the initial processing result before modification to generate a second expected processing result;

[0061] A reverse processing sample is generated according to the original data and the second expected processing result.

[0062] In an optional implementation, the page display submodule is specifically used for:

[0063] In response to a selection operation of a cell in the table, displaying second prompt information, wherein the second prompt information is used to prompt that artificial intelligence will be used for processing, and there are multiple pairs of repeated labels in the cells in the column where the cell is located;

[0064] In response to a confirmation operation of the artificial intelligence processing, an artificial intelligence setting page is displayed.

[0065] In an optional embodiment, the device further comprises:

[0066] The data and result display module is used to display the target original data in the table and the target processing results corresponding to the target original data on the preview result page;

[0067] Wherein, the target original data is the original data in the table processed by the user, and the target processing result is the multiple pairs of repeated labels;

[0068] A third prompt information display module, used to display third prompt information in the preview result page, wherein the third prompt information is used to prompt the initial artificial intelligence model to provide a processing sample;

[0069] An instruction triggering module, configured to trigger a sample collection instruction in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model;

[0070] A data and result acquisition module, configured to acquire the target original data and the target processing result corresponding to the target original data in response to a third sample collection instruction;

[0071] A third processing sample generating module, configured to generate a third processing sample according to the target original data and the target processing result corresponding to the target original data;

[0072] In the third model training module, Yang Hongyu trains the initial artificial intelligence model based on the third processed sample to obtain the target artificial intelligence model.

[0073] In an optional embodiment, the device further comprises:

[0074] An original data acquisition module, used for acquiring the original data in the table in response to the processing result generation operation;

[0075] A raw data processing module, used to process the raw data according to the label through the target artificial intelligence model to obtain a corresponding new processing result;

[0076] A result filling module is used to fill the new processing result corresponding to the original data in the processing result filling position.

[0077] In an optional embodiment, the device further comprises:

[0078] A page display module, used to display an instruction creation page in response to an instruction creation operation, wherein the instruction creation page is used to set a persona, example, and task in the instruction;

[0079] A character setting module, for determining a set character setting in response to a character setting operation, wherein the character setting is used to limit the working scope of the initial artificial intelligence model;

[0080] An example setting module, configured to determine, in response to an example setting operation, a set example, wherein the example is used to instruct the initial artificial intelligence model to perform data processing with reference to the example;

[0081] A task setting module, for determining a set task in response to a task setting operation, wherein the task is used to instruct the initial artificial intelligence model to process target raw data in the table;

[0082] An instruction providing module, configured to provide the persona, examples, and tasks in the instruction to the initial artificial intelligence model in response to a processing result generating operation;

[0083] A result filling module is used to fill the processing result filling position of the table with the processing result of the initial artificial intelligence model on the target original data in the table.

[0084] In an optional implementation, the page display module is specifically used to:

[0085] In response to the selection operation of artificial intelligence processing, an artificial intelligence setting page is displayed, wherein the artificial intelligence setting page is used to set the data source, the processing result filling position and the label;

[0086] In the artificial intelligence setting page, in response to a command center selection operation, a command center page is displayed;

[0087] On the command center page, in response to a command creation operation, displaying a command creation page;

[0088] The instruction providing module is specifically used for:

[0089] In response to the command confirmation operation, displaying the command center page;

[0090] On the command center page, in response to the confirmation operation, an artificial intelligence settings page is displayed;

[0091] In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

[0092] In an optional implementation, the page display module is specifically used to:

[0093] In response to the processing result preview operation, displaying the original data in the table and the initial processing result corresponding to the original data on the preview result page;

[0094] In the preview result page, in response to the instruction viewing operation, displaying an instruction details page;

[0095] On the instruction details page, in response to an instruction creation operation, displaying an instruction creation page;

[0096] The instruction providing module is specifically used for:

[0097] In response to the command confirmation operation, displaying the command center page;

[0098] On the command center page, in response to a confirmation operation, displaying the preview result page;

[0099] In response to a page jump operation, jumping from the preview result page to an artificial intelligence setting page;

[0100] In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

[0101] In an optional implementation, the example includes table data, labels, and output locations, and the example setting module is specifically used to:

[0102] In response to a setting operation of table data, determining example original data in the set table;

[0103] In response to a label setting operation, determining a label corresponding to the set example original data;

[0104] In response to the setting operation of the output position, a filling position corresponding to the set processing result of the example original data is determined.

[0105] In an optional implementation, the task includes table data, labels, and output locations, and the task setting module is specifically used to:

[0106] In response to a setting operation of table data, determining target original data in the set table;

[0107] In response to a tag setting operation, determining a tag corresponding to the set target original data;

[0108] In response to the setting operation of the output position, the set filling position of the processing result of the target original data is determined.

[0109] In a third aspect of the embodiments of the present application, there is further provided an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0110] Memory, used to store computer programs;

[0111] The processor is used to implement any sample collection method described in the first aspect when executing the program stored in the memory.

[0112] In a fourth aspect of the embodiments of the present application, a storage medium is further provided, wherein instructions are stored in the storage medium, and when the instructions are executed on a computer, the computer executes any sample collection method described in the first aspect above.

[0113] In a fifth aspect of the embodiments of the present application, a computer program product comprising instructions is also provided, which, when executed on a computer, enables the computer to execute any of the sample collection methods described above.

[0114] The technical solution provided by the embodiment of the present application obtains the original data in the table, obtains the initial processing results corresponding to the original data through the initial artificial intelligence model, and obtains the modified initial processing results in response to the modification operation on the initial processing results, wherein the modified initial processing results are used to provide the initial artificial intelligence model with the association relationship between the original data and the modified initial processing results to obtain the target artificial intelligence model. In this way, the modified initial processing results corresponding to the original data are collected and provided to the initial artificial intelligence model to learn the association relationship between the original data and the modified initial processing results. The artificial intelligence model learns the association relationship and subsequently refers to this association relationship for data processing, which will improve the accuracy of the processing results output by the artificial intelligence model. BRIEF DESCRIPTION OF THE DRAWINGS

[0115] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0116] 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, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0117] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0118] Figure 1 A schematic diagram of an implementation flow of a method for processing data in a table shown in an embodiment of the present application;

[0119] Figure 2 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0120] Figure 3This is a schematic diagram of displaying different artificial intelligence processing in a pop-up window in an embodiment of the present application;

[0121] Figure 4 A schematic diagram of an intelligent classification setting page shown in an embodiment of the present application;

[0122] Figure 5 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0123] Figure 6 A schematic diagram of performing a modification operation on an initial processing result shown in an embodiment of the present application;

[0124] Figure 7 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0125] Figure 8 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0126] Fig. 9 A schematic diagram of a sample processing configuration page shown in an embodiment of the present application;

[0127] Fig.10 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0128] Fig.11 This is a schematic diagram of prompting a user to use artificial intelligence processing shown in an embodiment of the present application;

[0129] Fig.12 A schematic diagram of a preview result page shown in an embodiment of the present application;

[0130] Fig.13 It is a schematic diagram of an implementation flow of another method for processing data in a table shown in an embodiment of the present application;

[0131] Fig.14 A schematic diagram of an instruction creation page shown in an embodiment of the present application;

[0132] Fig.15 It is a structural schematic diagram of a data processing device in a table shown in an embodiment of the present application;

[0133] Fig.16 It is a schematic diagram of the structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0134] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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 making creative work are within the scope of protection of this application.

[0135] The disclosure below provides many different embodiments or examples to realize the different structures of the present application. In order to simplify the disclosure of the present application, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present application. In addition, the present application can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0136] At present, since the user's table data cannot completely match the table data involved in the pre-training of the artificial intelligence model, the processing results output by the artificial intelligence model may be inaccurate. For this reason, the modified initial processing results can be collected, wherein the initial processing results corresponding to the original data in the table obtained by the initial artificial intelligence model are modified to obtain the modified initial processing results, which can be provided to the initial artificial intelligence model for learning, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0137] Based on this, Figure 1 FIG. 1 is a schematic diagram of an implementation flow of a method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0138] S101, obtaining original data in a table.

[0139] In the embodiment of the present application, when a user opens a spreadsheet document using spreadsheet processing software, the original data in the table of the spreadsheet document can be obtained. The original data can be any type of original data, and it can be the original data of one or more columns in the table. For example, the original data can be Figure 3 The original data can be the feedback content (column C) in the table, or the original data of one or more rows in the table, or the original data of some cells.

[0140] S102, obtaining an initial processing result corresponding to the original data through an initial artificial intelligence model.

[0141] In an embodiment of the present application, for the original data in the table, the initial processing results corresponding to the original data can be obtained through the initial artificial intelligence model, which means that the original data in the table is processed using the initial artificial intelligence model to obtain the corresponding initial processing results.

[0142] For example, for the original data in the table, classification is performed in the intelligent classification scenario. At this time, the initial classification results corresponding to the original data can be obtained through the initial artificial intelligence model, that is, the original data will have corresponding classification category labels.

[0143] It should be noted that the artificial intelligence model (i.e., the above-mentioned initial artificial intelligence model and the subsequent target artificial intelligence model) can be any neural network model, such as a large language model, and the large language model can be any existing large language model, which is not limited in the embodiments of the present application.

[0144] S103, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0145] In an embodiment of the present application, there may be erroneous processing results in the initial processing results, which means that there may be erroneous processing results in the initial processing results corresponding to the original data obtained through the initial artificial intelligence model. At this time, the user will actively modify it.

[0146] Of course, the embodiment of the present application can provide an error correction artificial intelligence model, which can identify the initial processing results, automatically determine the erroneous processing results in the initial processing results, and then automatically modify them.

[0147] Based on this, the user or the error correction artificial intelligence model performs a modification operation on the initial processing result, so that in response to the modification operation on the initial processing result, a modified initial processing result can be obtained, where the modified initial processing result is the correct processing result corresponding to the original data in the table.

[0148] Therefore, the modified initial processing results can be provided to the initial artificial intelligence model to learn the correlation between the original data and the modified initial processing results to obtain the target artificial intelligence model. The subsequent target artificial intelligence model will refer to this correlation to perform data processing, which will improve the accuracy of the processing results output by the artificial intelligence model.

[0149] Through the above description of the technical solution provided in the embodiment of the present application, the original data in the table is obtained, the initial processing results corresponding to the original data are obtained through the initial artificial intelligence model, and in response to the modification operation on the initial processing results, the modified initial processing results are obtained, wherein the modified initial processing results are provided to the initial artificial intelligence model to learn the association between the original data and the modified initial processing results, so as to obtain the target artificial intelligence model.

[0150] By obtaining the original data in the table, the initial processing results corresponding to the original data are obtained through the initial artificial intelligence model, and in response to the modification operation on the initial processing results, the modified initial processing results are obtained. The modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results, which will improve the accuracy of the processing results output by the artificial intelligence model.

[0151] For example, in the intelligent classification scenario, the original data in the table is obtained (XX membership expires...), and then the initial artificial intelligence model obtains the initial classification category label corresponding to the original data (such as purchase experience), and in response to the modification operation on the initial classification category label, the modified initial classification category label (such as consultation) is obtained. At this time, the modified initial classification category label can be provided to the initial artificial intelligence model to learn the association between the original data and the modified initial classification category label to obtain the target artificial intelligence model. Among them, the target artificial intelligence model will refer to this association for data processing, so as to improve the accuracy of the processing results output by the artificial intelligence model.

[0152] like Figure 2 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0153] S201, obtaining a table, and displaying an artificial intelligence setting page in response to a confirmation operation of artificial intelligence processing.

[0154] Among them, the artificial intelligence settings page is used to set the data source, processing result filling location and label.

[0155] S202, in response to the setting operation of the data source, determining the original data in the set table.

[0156] S203: In response to the tag setting operation, determine the set tag.

[0157] S204, in response to the setting operation of the processing result filling position, determining the set processing result filling position.

[0158] In the embodiment of the present application, when a user uses a spreadsheet processing software to open a spreadsheet document, a table in the spreadsheet document can be obtained, and the user can select a column of raw data in the table, and various artificial intelligence processing will be displayed in the form of a pop-up window, such as intelligent classification, intelligent extraction, etc. The user can select a certain artificial intelligence processing, and the confirmation operation of the artificial intelligence processing is triggered, so that in response to the confirmation operation of the artificial intelligence processing, the artificial intelligence setting page is displayed.

[0159] Among them, the artificial intelligence setting page is used to set the data source, the processing result filling position and the label. In the artificial intelligence setting page, the user can set the data source, thereby determining the original data in the set table in response to the data source setting operation, the user can set the label, thereby determining the set label in response to the label setting operation, and the user can also set the processing result filling position, thereby determining the set processing result filling position in response to the processing result filling position setting operation.

[0160] It should be noted that the labels here vary depending on the scenario. For example, in the intelligent classification scenario, it may include classification category labels, such as "consultation", "purchase experience" and other classification category labels, and in the intelligent extraction scenario, it may include extraction labels, such as "street name" and other extraction labels, which means that based on this extraction label, the specific street name is extracted from the original data, and in the intelligent emotion scenario, it can be an emotion label, such as "like" and other emotion labels.

[0161] For example, when a user uses spreadsheet processing software to open a spreadsheet document, the table in the spreadsheet document can be obtained. The user can select a column of raw data in the table, and various artificial intelligence processing functions will be displayed in the form of pop-up windows, such as Figure 3 As shown. At this time, the user can select the AI ​​processing function of smart classification, which will trigger the confirmation operation of smart classification, so that in response to the confirmation operation of smart classification, the smart classification setting page can be displayed. The smart classification setting page is used to set the data source, the processing result filling position and the classification category label. The user can set the data source, the processing result filling position and the classification category label, so as to respond to the setting operation of the data source, determine the original data in the set table (feedback content (column C)), respond to the setting operation of the label, determine the set classification category label (purchase experience, still prompt to pay, consultation), and respond to the setting operation of the processing result filling position. Determine the set processing result filling position (right side of column C), as shown Figure 4 shown.

[0162] S205, in response to the preset operation of the processing result, obtaining the original data in the table.

[0163] S206, using the initial artificial intelligence model, processing the original data according to the labels to obtain corresponding initial processing results.

[0164] S207, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0165] In an embodiment of the present application, after completing the setting operations in the above-mentioned artificial intelligence setting page, the user can perform preset operations on the processing results (such as preview operations or generation operations). At this time, the original data in the table will be obtained according to the above-mentioned setting operations of the data source.

[0166] In this way, the initial artificial intelligence model can be used to process the original data according to the set labels to obtain the corresponding initial processing results, and in response to the modification operation on the initial processing results, the modified initial processing results can be obtained.

[0167] Among them, the modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results to obtain the target artificial intelligence model.

[0168] It should be noted that, for the initial artificial intelligence model, the original data is processed according to the set labels to obtain the corresponding initial processing results, and the initial processing results here are filled into the corresponding positions in the table according to the set filling positions. Moreover, in different scenarios, for the initial artificial intelligence model, the original data is processed differently according to the set labels to obtain corresponding different initial processing results.

[0169] For example, in an intelligent classification scenario, for the initial artificial intelligence model, the original data in the set table is matched with the set classification category labels, so as to find the classification category labels that match the original data in the set table as the initial processing result of the original data in the set table, and the classification category labels that match the original data will be filled in a column to the right of the original data, thereby realizing intelligent classification of the original data in the table.

[0170] For example, in the intelligent extraction scenario, for the initial artificial intelligence model, the original data in the set table will be intelligently extracted according to the set extraction labels (such as extraction labels such as "street name", which means that based on this extraction label, specific street name and other partial data are extracted from the original data), and partial data of the original data in the set table will be extracted as the initial processing result of the original data in the set table, and the extracted partial data will be filled in a column to the right of the original data, thereby realizing the intelligent extraction of the original data in the table.

[0171] Through the above description of the technical solution provided in the embodiment of the present application, a table is obtained, and in response to the confirmation operation of the artificial intelligence processing, the artificial intelligence setting page is displayed, and the data source, the processing result filling position and the label are set on the artificial intelligence setting page, so as to respond to the preset operation of the processing result, the original data in the table is obtained, and the initial artificial intelligence model is used to process the original data according to the set label to obtain the corresponding initial processing result, and in response to the modification operation on the initial processing result, the modified initial processing result is obtained, and the modified initial processing result is used to provide the initial artificial intelligence model with the correlation relationship between the original data and the modified initial processing result, which will improve the accuracy of the processing result output by the artificial intelligence model.

[0172] In addition, for the original data in the table, the initial artificial intelligence model is used to process it in a specific scenario and output the initial processing results. A preview function of the initial processing results can be provided here. When the user previews the initial processing results, it is possible to modify the initial processing results (for example, an erroneous initial processing result) output by the initial artificial intelligence model, so as to obtain the modified initial processing results. The modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results, so as to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0173] Based on this, Figure 5 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0174] S501, obtaining a table, and displaying an artificial intelligence setting page in response to a confirmation operation of artificial intelligence processing.

[0175] In the embodiment of the present application, this step is similar to the above-mentioned step S201, and the embodiment of the present application will not be described one by one here.

[0176] Among them, the artificial intelligence settings page is used to set the data source, processing result filling location and label.

[0177] S502, in response to the setting operation of the data source, determining the original data in the set table.

[0178] In the embodiment of the present application, this step is similar to the above-mentioned step S202, and the embodiment of the present application will not be described one by one here.

[0179] S503: In response to the tag setting operation, determine the set tag.

[0180] In the embodiment of the present application, this step is similar to the above-mentioned step S203, and the embodiment of the present application will not be described one by one here.

[0181] S504, in response to the setting operation of the processing result filling position, determining the set processing result filling position.

[0182] In the embodiment of the present application, this step is similar to the above-mentioned step S204, and the embodiment of the present application will not be described one by one here.

[0183] S505: In response to the processing result preview operation, original data in the table is obtained.

[0184] S506, using the initial artificial intelligence model, processing the original data according to the labels to obtain the corresponding initial processing results.

[0185] S507, displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page.

[0186] S508, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0187] In the embodiment of the present application, after the above setting operation on the above artificial intelligence setting page, the user can perform a processing result preview operation. Figure 4 The smart classification setting page shown has a "Preview processing results" button. Clicking this button can trigger the processing result preview operation. In response to the processing result preview operation, on the one hand, the original data in the table is obtained, and the initial artificial intelligence model is used to process the original data according to the labels to obtain the corresponding initial processing results. On the other hand, it will jump to the preview result page, at which time the original data in the table and the initial processing results corresponding to the original data are displayed on the preview result page.

[0188] In addition, part of the original data in the table and the initial processing results corresponding to part of the original data can be displayed on the preview result page, and the preview range can also be modified to adjust part of the original data in the table displayed on the preview result page and the initial processing results corresponding to part of the original data.

[0189] Among them, the initial processing results corresponding to the original data displayed in the preview result page may contain erroneous initial processing results. At this time, the user can perform modification operations on the processing results, such as Figure 6 As shown, in response to the modification operation on the initial processing result, a modified initial processing result can be obtained. The modified initial processing result is used to provide the initial artificial intelligence model with the association relationship between the original data and the modified initial processing result to obtain the target artificial intelligence model.

[0190] In this way, the original data in the table is processed using the initial artificial intelligence model in a specific scenario, and the initial processing results are output, and a preview function of the initial processing results is provided. When the user previews the initial processing results, it is possible to modify the initial processing results (for example, erroneous initial processing results) output by the initial artificial intelligence model, so that the modified initial processing results can be obtained. The modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results, so as to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0191] In addition, for the original data in the table, the initial artificial intelligence model is used to process it in a specific scenario, and the initial processing results are output, and the function of generating the initial processing results is provided. In this way, the processing result filling position of the table will be filled with the initial processing results output by the initial artificial intelligence model. The user may modify the initial processing results output by the initial artificial intelligence model (for example, an erroneous initial processing result), so as to obtain the modified initial processing results. The modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results, so as to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0192] Based on this, Figure 7 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0193] S701, obtaining a table, and displaying an artificial intelligence setting page in response to a confirmation operation of the artificial intelligence processing.

[0194] In the embodiment of the present application, this step is similar to the above-mentioned step S201, and the embodiment of the present application will not be described one by one here.

[0195] Among them, the artificial intelligence settings page is used to set the data source, processing result filling location and label.

[0196] S702, in response to the setting operation of the data source, determining the original data in the set table.

[0197] In the embodiment of the present application, this step is similar to the above-mentioned step S202, and the embodiment of the present application will not be described one by one here.

[0198] S703: In response to the tag setting operation, determine the set tag.

[0199] In the embodiment of the present application, this step is similar to the above-mentioned step S203, and the embodiment of the present application will not be described one by one here.

[0200] S704, in response to the setting operation of the processing result filling position, determining the set processing result filling position.

[0201] In the embodiment of the present application, this step is similar to the above-mentioned step S204, and the embodiment of the present application will not be described one by one here.

[0202] S705, in response to the processing result generating operation, obtaining original data in the table.

[0203] S706, using the initial artificial intelligence model, processing the original data according to the labels to obtain the corresponding initial processing results.

[0204] S707, filling the initial processing result corresponding to the original data in the processing result filling position.

[0205] S708, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0206] In the embodiment of the present application, after the above setting operation of the above artificial intelligence setting page, the user can perform the processing result generation operation. Figure 4 The smart classification setting page shown has a "Start Processing" button. Clicking this button can trigger the processing result generation operation. In response to the processing result generation operation, on the one hand, the original data in the table is obtained, and the initial artificial intelligence model is used to process the original data according to the labels to obtain the corresponding initial processing results. On the other hand, the initial processing results corresponding to the original data are filled in the processing result filling position of the table, for example, the initial processing results corresponding to the original data are filled on the right side of column C of the table.

[0207] Among them, for the initial processing result corresponding to the original data filled in the processing result filling position of the table, there may be an erroneous initial processing result. At this time, the user can perform a modification operation on the processing result, so that the modified initial processing result can be obtained in response to the modification operation on the initial processing result. Among them, the modified initial processing result is used to provide the initial artificial intelligence model with the association relationship between the original data and the modified initial processing result to obtain the target artificial intelligence model.

[0208] In this way, for the original data in the table, the initial artificial intelligence model is used to process it in a specific scenario, and the initial processing results are output, and the function of generating the initial processing results is provided. In this way, the processing result filling position of the table will be filled with the initial processing results output by the initial artificial intelligence model. The user may modify the initial processing results output by the initial artificial intelligence model (for example, an erroneous initial processing result), so as to obtain the modified initial processing results. The modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results, so as to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0209] In addition, for the original data in the table, the initial artificial intelligence model is used to process it in a specific scenario and output the initial processing results. The user may modify the initial processing results (for example, incorrect initial processing results) output by the initial artificial intelligence model. At this time, processing samples can be collected for training the initial artificial intelligence model to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0210] Based on this, Figure 8 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0211] S801, obtaining original data in the table.

[0212] In the embodiment of the present application, this step is similar to the above-mentioned step S101, and the embodiment of the present application will not be described one by one here.

[0213] S802, obtaining an initial processing result corresponding to the original data through an initial artificial intelligence model.

[0214] In the embodiment of the present application, this step is similar to the above-mentioned step S102, and the embodiment of the present application will not be described one by one here.

[0215] S803, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0216] In the embodiment of the present application, this step is similar to the above-mentioned step S103, and the embodiment of the present application will not be described one by one here.

[0217] Among them, the modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results to obtain the target artificial intelligence model.

[0218] S804, displaying a first prompt message, where the first prompt message is used to prompt the initial artificial intelligence model to provide a processing sample.

[0219] S805, in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction.

[0220] S806 , in response to the first sample collection instruction, generate a first processed sample according to the original data and the modified initial processing result.

[0221] S807: Train the initial artificial intelligence model based on the first processed sample to obtain a target artificial intelligence model.

[0222] In an embodiment of the present application, in response to a modification operation on the initial processing result, after obtaining the modified initial processing result, a first prompt information may be displayed, where the first prompt information may be, for example, "provide reference to AI", and the first prompt information is used to prompt to provide processing samples to the initial artificial intelligence model. If the user confirms to provide the processing sample to the initial artificial intelligence model, a confirmation operation may be performed, thereby triggering a sample collection instruction in response to the confirmation operation of providing the processing sample to the initial artificial intelligence model, where the sample collection instruction includes a first sample collection instruction, so that in response to the first sample collection instruction, a first processing sample may be generated according to the original data and the modified initial processing result, and the initial artificial intelligence model may be trained based on the first processing sample to obtain a target artificial intelligence model.

[0223] It should be noted that if Figure 6 The button "Provide reference to AI" in the preview result page shown is a button in itself, and can also serve as the role of the first prompt information. At this time, the user can use this button to perform the confirmation operation of providing the processing sample to the initial artificial intelligence model. Based on this, the above steps S801 to S803 can specifically be the above steps S501 to S508, and then there is a button "Provide reference to AI" in the preview result page. This button also displays the first prompt information of "Provide reference to AI". At this time, the user can click this button to provide the initial artificial intelligence model with the execution confirmation operation of the processing sample, thereby responding to the confirmation operation of providing the processing sample to the initial artificial intelligence model, triggering the sample collection instruction, where the sample collection instruction includes the first sample collection instruction, so that in response to the first sample collection instruction, the first processing sample can be generated according to the original data and the modified initial processing result, and the initial artificial intelligence model is trained based on the first processing sample to obtain the target artificial intelligence model.

[0224] Among them, for the above-mentioned first processing sample, the first processing sample here can be a forward processing sample or a reverse processing sample. Based on this, the modified initial processing result is determined as the first expected processing result corresponding to the original data, and the forward processing sample is generated according to the original data and the first expected processing result corresponding to the original data. Alternatively, the initial processing result before the modification corresponding to the modified initial processing result is determined, and the preset reverse prompt information is combined with the initial processing result before the modification to generate a second expected processing result, and the reverse processing sample is generated according to the original data and the second expected processing result.

[0225] For example, the initial processing result corresponding to the original data is a, and the modified initial processing result is a', which means that the modified initial processing result is the correct processing result. At this time, the modified initial processing result can be determined as the first expected processing result corresponding to the original data, thereby generating a forward processing sample based on the original data and the first expected processing result corresponding to the original data.

[0226] For example, the initial processing result after modification is a', and the initial processing result before modification is a. Determine the initial processing result a before modification corresponding to the initial processing result a' after modification, and combine the preset reverse prompt information of "Do not process as" with the initial processing result a before modification to generate a second expected processing result, such as "Do not process as a". Generate a reverse processing sample based on the original data and the second expected processing result.

[0227] In addition, in an embodiment of the present application, in response to a confirmation operation of providing a processed sample to the initial artificial intelligence model and triggering a sample collection instruction, the sample collection instruction here also includes a second sample collection instruction. In response to the second sample collection instruction, a processing sample configuration page is displayed, and the processing sample configuration page is used to configure a second processing sample. On the processing sample configuration page, the target original data in the table input by the user and the target processing result corresponding to the target original data are obtained. According to the target original data and the target processing result corresponding to the target original data, a second processing sample is generated, and the initial artificial intelligence model is trained based on the second processing sample to obtain the target artificial intelligence model.

[0228] It should be noted that for forward processing samples, the initial artificial intelligence model is trained based on the forward processing samples to obtain the target artificial intelligence model. The target artificial intelligence model can learn that when processing the original data, it can be processed as the modified initial processing result, while for reverse processing samples, the initial artificial intelligence model is trained based on the reverse processing samples to obtain the target artificial intelligence model. The target artificial intelligence model can learn that when processing the original data, it cannot be processed as the initial processing result before modification. In this way, when the target artificial intelligence model is used for data processing subsequently, if data similar or identical to the original data is encountered, it will be processed as the modified initial processing result, and avoid being processed as the initial processing result before modification.

[0229] For example, the above steps S801 to S803 may specifically be the above steps S501 to S508, and then there is a button "Provide reference to AI" in the preview result page. This button also displays the first prompt information of "Provide reference to AI". At this time, the user can click the button to perform the confirmation operation of providing the processed samples to the initial artificial intelligence model, thereby triggering the sample collection instruction in response to the confirmation operation of providing the processed samples to the initial artificial intelligence model. The sample collection instruction here includes not only the first sample collection instruction, but also the second sample collection instruction, thereby displaying the processing sample configuration page in response to the second sample collection instruction. The processing sample configuration page is used to configure the second processing sample, which means that the user can configure the second processing sample by himself. On the processing sample configuration page, the target original data in the form entered by the user is obtained (XX membership will be purchased separately after expiration...), as well as the target processing result corresponding to the target original data (still prompting for payment), such as Fig. 9 shown.

[0230] Among them, the target processing result can generally be a first expected processing result in the positive direction, and a positive processing sample can be obtained. Based on this, the target original data in the table input by the user and the first expected processing result corresponding to the target original data are obtained, and the first expected processing result contains the correct processing result. According to the target original data and the first expected processing result corresponding to the target original data, a positive processing sample is generated.

[0231] In addition, for the target processing result, a reverse second expected processing result can also be obtained, such as "do not process as xxx", to obtain a reverse processing sample. Based on this, the target original data in the table input by the user and the second expected processing result corresponding to the target original data are obtained. Compared with the first expected processing result, the second expected processing result contains reverse prompt information and contains an incorrect processing result. According to the target original data and the second expected processing result corresponding to the target original data, a reverse processing sample is generated.

[0232] In this way, for the original data in the table, the initial artificial intelligence model is used to process it in a specific scenario and output the initial processing results. The user may modify the initial processing results (for example, incorrect initial processing results) output by the initial artificial intelligence model. At this time, processing samples can be collected for training the initial artificial intelligence model to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0233] In addition, for the raw data in the user table, the user can process the target raw data correctly by himself. At this time, the user can be guided to provide processing samples based on the correctly processed target raw data for training the initial artificial intelligence model to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model. Based on this, Fig.10 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0234] S1001, obtaining a table, and in response to a selection operation of a cell in the table, displaying a second prompt message, wherein the second prompt message is used to prompt that artificial intelligence will be used for processing, and there are multiple pairs of repeated labels in the cells in the column where the cell is located.

[0235] In an embodiment of the present application, when a user uses a spreadsheet software to open a spreadsheet document, the table in the spreadsheet document can be obtained and then displayed in the view area of ​​the spreadsheet processing software. At this time, the user can process the original data in the table by himself. The processing here refers to the user matching appropriate labels to the original data by himself.

[0236] In the process of processing raw data, the user will select a cell and fill in the label that matches the raw data. If there are multiple pairs of repeated labels in the cells in the column where the cell is located, it means that the user's processing of the raw data can be replaced by artificial intelligence processing. In response to the user's cell selection operation in the table, a second prompt message can be displayed, where the second prompt message is used to prompt that artificial intelligence will be used for processing, and the confirmation operation of the artificial intelligence processing can be triggered according to the prompt.

[0237] For example, when a user opens a spreadsheet document using spreadsheet processing software, the table in the spreadsheet document can be obtained and then displayed in the view area of ​​the spreadsheet processing software. The user can classify the raw data in the user's table by himself. Classification here means that the user matches the raw data with appropriate classification category labels. In the process of classifying raw data, the user will select a cell and then fill in the classification category label corresponding to the raw data. For the cells in the column where the cell is located, if there are multiple pairs of repeated classification category labels, such as Fig.11 The "still prompt to pay", "consultation" and "purchase experience" shown indicate that the user's processing of the original data can be replaced by intelligent classification. At that time, it was an intelligent classification scenario, so in response to the user's selection operation of a cell in the table, a second prompt message can be displayed. The second prompt message is used to prompt that intelligent classification will be used for processing. At this time, the user clicks the "Set" button, which will trigger the confirmation operation of the intelligent classification.

[0238] It should be noted that if there are multiple pairs of repeated labels in the cells of the column where the cell is located, it means that the user has processed some target raw data in the raw data, and each target raw data has a corresponding processing result, which is the label set by the user for the target raw data, such as Fig.11 "Still prompt to pay", "Consultation", "Purchase experience" as shown.

[0239] S1002, in response to a confirmation operation of the artificial intelligence processing, displaying an artificial intelligence setting page.

[0240] In the embodiment of the present application, this step is similar to the above-mentioned step S201, and the embodiment of the present application will not be described one by one here.

[0241] Among them, the artificial intelligence settings page is used to set the data source, processing result filling location and label.

[0242] S1003, in response to the setting operation of the data source, determining the original data in the set table.

[0243] In the embodiment of the present application, this step is similar to the above-mentioned step S202, and the embodiment of the present application will not be described one by one here.

[0244] S1004: In response to the tag setting operation, determine the set tag.

[0245] In the embodiment of the present application, this step is similar to the above-mentioned step S203, and the embodiment of the present application will not be described one by one here.

[0246] S1005 , in response to the setting operation of the processing result filling position, determining the set processing result filling position.

[0247] In the embodiment of the present application, this step is similar to the above-mentioned step S204, and the embodiment of the present application will not be described one by one here.

[0248] S1006, in response to the processing result preview operation, obtaining original data in the table.

[0249] In the embodiment of the present application, this step is similar to the above-mentioned step S505, and the embodiment of the present application will not be described one by one here.

[0250] S1007, using the initial artificial intelligence model, process the original data according to the labels to obtain the corresponding initial processing results.

[0251] In the embodiment of the present application, this step is similar to the above-mentioned step S206, and the embodiment of the present application will not be described in detail here.

[0252] S1008, displaying the target original data in the table and the target processing results corresponding to the target original data on the preview result page.

[0253] In the embodiment of the present application, after the above setting operation on the above artificial intelligence setting page, the user can perform a processing result preview operation. Figure 4 The smart classification setting page shown has a "Preview Processing Results" button. Clicking this button can trigger the processing result preview operation. In response to the processing result preview operation, on the one hand, the original data in the table is obtained, and the original data is processed according to the labels using the initial artificial intelligence model to obtain the corresponding initial processing results. On the other hand, it will jump to the preview result page, at which time the target original data in the table and the target processing results corresponding to the target original data are displayed on the preview result page. Among them, the target original data is the original data in the table processed by the user, and the target processing results are the above-mentioned multiple pairs of repeated labels.

[0254] It should be noted that when the user previews the processing results, the preview result page will by default display the target original data in the table and the target processing results corresponding to the target original data. The preview range can be adjusted later so that the original data in the table and the initial processing results corresponding to the original data can be displayed on the preview result page.

[0255] S1009, in the preview result page, a third prompt information is displayed, and the third prompt information is used to prompt to provide processing samples to the initial artificial intelligence model.

[0256] S1010, in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction.

[0257] S1011, in response to a third sample collection instruction, acquiring target original data and a target processing result corresponding to the target original data.

[0258] S1012, generating a third processed sample according to the target original data and the target processing result corresponding to the target original data.

[0259] S1013, training the initial artificial intelligence model based on the third processed sample to obtain a target artificial intelligence model.

[0260] In the embodiment of the present application, a third prompt message is displayed on the preview result page, and the third prompt message is used to prompt the initial artificial intelligence model to provide a processing sample. The third prompt message can be, for example, "one-click reference, more accurate results", such as Fig.12 It should be noted that for Fig.12 The third prompt message shown, "Provide reference with one click, and the results are more accurate", is also a button. Clicking the button can trigger the confirmation operation of providing the processed samples to the initial artificial intelligence model.

[0261] Based on this, in response to the confirmation operation of providing the processing sample to the initial artificial intelligence model, the sample collection instruction is triggered. The sample collection instruction here includes a third sample collection instruction. In response to the third sample collection instruction, the target original data and the target processing result corresponding to the target original data are obtained. According to the target original data and the target processing result corresponding to the target original data, a third processed sample is generated. The initial artificial intelligence model is trained based on the third processed sample to obtain the target artificial intelligence model.

[0262] S1014, displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page.

[0263] In an embodiment of the present application, the user can adjust the preview range. At this time, the original data in the table and the initial processing results corresponding to the original data can be displayed on the preview result page. The original data here may not include the above-mentioned target original data (the original data processed by the user himself), and the corresponding initial processing results may not include the above-mentioned target processing results (that is, the above-mentioned multiple pairs of repeated labels).

[0264] S1015, in response to the modification operation on the initial processing result, obtaining a modified initial processing result.

[0265] In the embodiment of the present application, this step is similar to the above-mentioned step S103, and the embodiment of the present application will not be described one by one here.

[0266] Among them, the modified initial processing results are used to provide the initial artificial intelligence model with learning the correlation between the original data and the modified initial processing results to obtain the target artificial intelligence model.

[0267] In this way, for the original data in the user table, the user can correctly process the target original data by himself. At this time, the user can be guided to provide processing samples based on the correctly processed target original data for training the initial artificial intelligence model to obtain the target artificial intelligence model, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0268] In addition, after the above processing, the latest target artificial intelligence model can be obtained, and the target artificial intelligence model can be used for data processing. Thus, the user can perform the processing result generation operation to obtain the original data in the table. The original data here can be the new original data in the table, that is, the table data in the table that has never generated a processing result, and of course, it can also be the table data that has generated a processing result. Through the target artificial intelligence model, the original data is processed according to the label to obtain the corresponding new processing result, and the new processing result corresponding to the original data is filled in the processing result filling position of the table.

[0269] In addition, for the initial artificial intelligence model, the means of interaction with the initial artificial intelligence model is usually prompts. The richer and more regular prompts can effectively improve the accuracy of the processing results output by the initial artificial intelligence model. Based on this consideration, the embodiment of the present application structures the various parts of the prompts and divides them into three modules: persona, example, and task. These can be used as input for the initial artificial intelligence model to use as constraints, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0270] Based on this, Fig.13 FIG. 1 is a schematic diagram of an implementation flow of another method for processing data in a table provided in an embodiment of the present application. The method is applied to an electronic device and may specifically include the following steps:

[0271] S1301, in response to an instruction creation operation, displaying an instruction creation page, where the instruction creation page is used to set the persona, examples, and tasks in the instruction.

[0272] In an embodiment of the present application, a user can trigger an instruction creation operation, and thus, in response to the instruction creation operation, an instruction creation page can be displayed. The instruction creation page is used to set the character, examples, and tasks in the instructions. The instructions here are prompt words, which means that the prompt words are provided to the user in the form of instructions, allowing the user to set the three major modules of character, examples, and tasks by themselves.

[0273] S1302, in response to a personality setting operation, determining the set personality, where the personality is used to limit the working scope of the initial artificial intelligence model.

[0274] In an embodiment of the present application, for the personality in the instruction, the user can perform the setting operation by himself, thereby determining the set personality in response to the personality setting operation, and the personality is used to limit the working scope of the initial artificial intelligence model.

[0275] For example, for the setting of the character, it can be "I will provide you with a table data and a set of labels, please select the most matching label based on the table data", which can limit the working scope of the initial artificial intelligence model.

[0276] S1303, in response to the example setting operation, determining the set example, where the example is used to instruct the initial artificial intelligence model to refer to the example for data processing.

[0277] In an embodiment of the present application, for the examples in the instructions, the user can perform the setting operation by himself, thereby determining the set examples in response to the example setting operation, and the examples are used to instruct the initial artificial intelligence model to refer to the examples for data processing.

[0278] Examples generally include table data, labels, and output locations, and users can perform setting operations on the table data, labels, output locations, etc. respectively.

[0279] Thus, in response to the setting operation of the table data, the example original data in the set table can be determined, in response to the setting operation of the label, the label corresponding to the set example original data can be determined, and in response to the setting operation of the output position, the filling position corresponding to the processing result of the set example original data can be determined, such as Fig.14 shown.

[0280] S1304, in response to the task setting operation, determining the set task, where the task is used to instruct the initial artificial intelligence model to process the target original data in the table.

[0281] In an embodiment of the present application, for the tasks in the instructions, the user can perform the setting operation by himself, thereby determining the set task in response to the task setting operation, and the task is used to instruct the initial artificial intelligence model to process the target original data in the table.

[0282] Among them, tasks are similar to examples, usually including table data, labels, and output locations. Users can perform setting operations on the table data, labels, output locations, etc. respectively.

[0283] Thus, in response to the setting operation of the table data, the target original data in the set table is determined, in response to the setting operation of the label, the label corresponding to the set target original data is determined, and in response to the setting operation of the output position, the filling position of the processing result of the set target original data is determined, such as Fig.14 shown.

[0284] S1305, in response to the processing result generation operation, providing the persona, examples, and tasks in the instruction to the initial artificial intelligence model.

[0285] S1306, filling the processing result of the initial artificial intelligence model on the target original data in the table in the processing result filling position of the table.

[0286] In an embodiment of the present application, after the above-mentioned setting operation, the character settings, examples, and tasks in the instruction are set. At this time, the instruction can be used as a constraint condition of the initial artificial intelligence model for data processing, thereby responding to the processing result generation operation, and providing the character settings, examples, and tasks in the instruction to the initial artificial intelligence model, and filling the processing result filling position of the table with the processing results of the initial artificial intelligence model on the target original data in the table.

[0287] It should be noted that the filling position of the processing result of the table here refers to the filling position set in the above task, and the processing result here can be one of the labels set above, or it can be the result of processing the target original data according to the label. The embodiment of the present application is not limited to this.

[0288] In this way, the various parts of the prompt word are structured and divided into three modules: character setting, examples, and tasks. These can be used as input for the initial artificial intelligence model as constraints, thereby improving the accuracy of the processing results output by the artificial intelligence model.

[0289] In addition, the timing of creating instructions can be varied, for example, instructions can be created directly in the instruction center. For the above step S1301, the user can use the spreadsheet processing software to open the spreadsheet document, and the table in the spreadsheet document is displayed in the view area of ​​the spreadsheet processing software. At this time, the user can select artificial intelligence processing to process the table data. At this time, in response to the selection operation of artificial intelligence processing, the artificial intelligence setting page can be displayed, and the artificial intelligence setting page is used to set the data source, the processing result filling position and the label.

[0290] For the AI ​​settings page, for example Figure 4As shown, there is a "Command Center" button, and the user can click the button to jump to the command center page. Therefore, the user clicks the button, triggering the command center selection operation, so that the command center page can be displayed in response to the command center selection operation, and on the command center page, in response to the command creation operation, the command creation page is displayed.

[0291] For the above step S1305, specifically, after creating the instruction, the user can confirm the created instruction in the instruction center, thereby triggering the instruction confirmation operation. In response to the instruction confirmation operation, the instruction center page is displayed. On the instruction center page, in response to the confirmation operation, the artificial intelligence setting page is displayed; on the artificial intelligence setting page, in response to the processing result generation operation, the character setting, examples and tasks in the instruction are provided to the initial artificial intelligence model.

[0292] In addition, the creation time of the instruction can be created by imitating other instructions. Based on this, the above step S1301 can specifically allow the user to perform a preview operation on the processing result. In response to the processing result preview operation, the original data in the table and the initial processing result corresponding to the original data are displayed on the preview result page, wherein the initial processing result is the result obtained by processing the original data using the initial artificial intelligence model.

[0293] For the preview result page, for example Figure 6 As shown, there is a "View instruction details" button, and the user can click the button to trigger the instruction viewing operation, so that in response to the instruction viewing operation, the instruction details page is displayed. For the instruction details page, there is a "Create a new instruction using this as a template" button, and the user can click the button to trigger the instruction creation operation, and in response to the instruction creation operation, the instruction creation page is displayed.

[0294] For the above step S1305, specifically, after creating the instruction, the user can confirm the created instruction in the instruction center, thereby triggering the instruction confirmation operation. In response to the instruction confirmation operation, the instruction center page is displayed. On the instruction center page, in response to the confirmation operation, a preview result page is displayed. In response to the page jump operation, the preview result page jumps to the artificial intelligence setting page. In the artificial intelligence setting page, in response to the processing result generation operation, the persona, examples and tasks in the instruction are provided to the initial artificial intelligence model.

[0295] Corresponding to the above method embodiment, the present application embodiment also provides a sample collection device, such as Fig.15 As shown, the device may include: a data acquisition module 1510 , a data processing module 1520 , and a result modification module 1530 .

[0296] The data acquisition module 1510 is used to acquire the original data in the table;

[0297] A data processing module 1520 is used to obtain an initial processing result corresponding to the original data through an initial artificial intelligence model;

[0298] A result modification module 1530, configured to obtain a modified initial processing result in response to a modification operation on the initial processing result;

[0299] Among them, the modified initial processing result is used to provide the initial artificial intelligence model with a learning relationship between the original data and the modified initial processing result to obtain a target artificial intelligence model.

[0300] In an optional implementation, the data acquisition module specifically includes:

[0301] a page display submodule, for acquiring a table and, in response to a confirmation operation of the artificial intelligence processing, displaying an artificial intelligence setting page;

[0302] The artificial intelligence setting page is used to set the data source, the processing result filling position and the label;

[0303] The original data setting submodule is used to determine the original data in the set table in response to the setting operation of the data source;

[0304] The label setting submodule is used to determine the set label in response to the label setting operation;

[0305] A filling position setting submodule, for determining a set processing result filling position in response to a setting operation of the processing result filling position;

[0306] A data acquisition submodule, used for acquiring the original data in the table in response to a preset operation of the processing result;

[0307] The data processing module is specifically used for:

[0308] The initial artificial intelligence model is used to process the raw data according to the labels to obtain corresponding initial processing results.

[0309] In an optional implementation, the data acquisition submodule is specifically used for:

[0310] In response to the processing result preview operation, obtaining the original data in the table;

[0311] The result modification module is specifically used for:

[0312] Displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page;

[0313] In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

[0314] In an optional implementation, the data acquisition submodule is specifically used for:

[0315] In response to the processing result generating operation, obtaining the original data in the table;

[0316] The result modification module is specifically used for:

[0317] Fill the initial processing result corresponding to the original data in the processing result filling position;

[0318] In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

[0319] In an optional embodiment, the device further comprises:

[0320] A first prompt information display module, used to display first prompt information, wherein the first prompt information is used to prompt the initial artificial intelligence model to provide a processing sample;

[0321] An instruction triggering module, configured to trigger a sample collection instruction in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model;

[0322] A first processed sample generating module, configured to generate a first processed sample according to the original data and the modified initial processing result in response to a first sample collecting instruction;

[0323] The first model training module is used to train the initial artificial intelligence model based on the first processed sample to obtain a target artificial intelligence model.

[0324] In an optional embodiment, the device further comprises:

[0325] a configuration page display module, configured to display a processing sample configuration page in response to the second sample collection instruction, wherein the processing sample configuration page is used to configure the second processing sample;

[0326] A data and result acquisition module, used to acquire the target original data in the table input by the user and the target processing result corresponding to the target original data on the processing sample configuration page;

[0327] A second processing sample generating module, configured to generate a second processing sample according to the target original data and a target processing result corresponding to the target original data;

[0328] The second model training module is used to train the initial artificial intelligence model based on the second processed sample to obtain a target artificial intelligence model.

[0329] In an optional implementation, the first processed sample generating module is specifically configured to:

[0330] Determining the modified initial processing result as the first expected processing result corresponding to the original data;

[0331] A forward processing sample is generated according to the original data and the first expected processing result corresponding to the original data.

[0332] In an optional implementation, the first processed sample generating module is specifically configured to:

[0333] Determine the initial processing result before modification corresponding to the initial processing result after modification;

[0334] Combining the preset reverse prompt information with the initial processing result before modification to generate a second expected processing result;

[0335] A reverse processing sample is generated according to the original data and the second expected processing result.

[0336] In an optional implementation, the page display submodule is specifically used for:

[0337] In response to a selection operation of a cell in the table, displaying second prompt information, wherein the second prompt information is used to prompt that artificial intelligence will be used for processing, and there are multiple pairs of repeated labels in the cells in the column where the cell is located;

[0338] In response to a confirmation operation of the artificial intelligence processing, an artificial intelligence setting page is displayed.

[0339] In an optional embodiment, the device further comprises:

[0340] The data and result display module is used to display the target original data in the table and the target processing results corresponding to the target original data on the preview result page;

[0341] Wherein, the target original data is the original data in the table processed by the user, and the target processing result is the multiple pairs of repeated labels;

[0342] A third prompt information display module, used to display third prompt information in the preview result page, wherein the third prompt information is used to prompt the initial artificial intelligence model to provide a processing sample;

[0343] An instruction triggering module, configured to trigger a sample collection instruction in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model;

[0344] A data and result acquisition module, configured to acquire the target original data and the target processing result corresponding to the target original data in response to a third sample collection instruction;

[0345] A third processing sample generating module, configured to generate a third processing sample according to the target original data and the target processing result corresponding to the target original data;

[0346] In the third model training module, Yang Hongyu trains the initial artificial intelligence model based on the third processed sample to obtain the target artificial intelligence model.

[0347] In an optional embodiment, the device further comprises:

[0348] An original data acquisition module, used for acquiring the original data in the table in response to the processing result generation operation;

[0349] A raw data processing module, used to process the raw data according to the label through the target artificial intelligence model to obtain a corresponding new processing result;

[0350] A result filling module is used to fill the new processing result corresponding to the original data in the processing result filling position.

[0351] In an optional embodiment, the device further comprises:

[0352] A page display module, used to display an instruction creation page in response to an instruction creation operation, wherein the instruction creation page is used to set a persona, example, and task in the instruction;

[0353] A character setting module, for determining a set character setting in response to a character setting operation, wherein the character setting is used to limit the working scope of the initial artificial intelligence model;

[0354] An example setting module, configured to determine, in response to an example setting operation, a set example, wherein the example is used to instruct the initial artificial intelligence model to perform data processing with reference to the example;

[0355] A task setting module, for determining a set task in response to a task setting operation, wherein the task is used to instruct the initial artificial intelligence model to process target raw data in the table;

[0356] An instruction providing module, configured to provide the persona, examples, and tasks in the instruction to the initial artificial intelligence model in response to a processing result generating operation;

[0357] A result filling module is used to fill the processing result filling position of the table with the processing result of the initial artificial intelligence model on the target original data in the table.

[0358] In an optional implementation, the page display module is specifically used to:

[0359] In response to the selection operation of artificial intelligence processing, an artificial intelligence setting page is displayed, wherein the artificial intelligence setting page is used to set the data source, the processing result filling position and the label;

[0360] In the artificial intelligence setting page, in response to a command center selection operation, a command center page is displayed;

[0361] On the command center page, in response to a command creation operation, displaying a command creation page;

[0362] The instruction providing module is specifically used for:

[0363] In response to the command confirmation operation, displaying the command center page;

[0364] On the command center page, in response to the confirmation operation, an artificial intelligence settings page is displayed;

[0365] In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

[0366] In an optional implementation, the page display module is specifically used to:

[0367] In response to the processing result preview operation, displaying the original data in the table and the initial processing result corresponding to the original data on the preview result page;

[0368] In the preview result page, in response to the instruction viewing operation, displaying an instruction details page;

[0369] On the instruction details page, in response to an instruction creation operation, displaying an instruction creation page;

[0370] The instruction providing module is specifically used for:

[0371] In response to the command confirmation operation, displaying the command center page;

[0372] On the command center page, in response to a confirmation operation, displaying the preview result page;

[0373] In response to a page jump operation, jumping from the preview result page to an artificial intelligence setting page;

[0374] In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

[0375] In an optional implementation, the example includes table data, labels, and output locations, and the example setting module is specifically used to:

[0376] In response to a setting operation of table data, determining example original data in the set table;

[0377] In response to a label setting operation, determining a label corresponding to the set example original data;

[0378] In response to the setting operation of the output position, a filling position corresponding to the set processing result of the example original data is determined.

[0379] In an optional implementation, the task includes table data, labels, and output locations, and the task setting module is specifically used to:

[0380] In response to a setting operation of table data, determining target original data in the set table;

[0381] In response to a tag setting operation, determining a tag corresponding to the set target original data;

[0382] In response to the setting operation of the output position, the set filling position of the processing result of the target original data is determined.

[0383] The present application also provides an electronic device, such as Fig.16 As shown, it includes a processor 161, a communication interface 162, a memory 163 and a communication bus 164, wherein the processor 161, the communication interface 162, and the memory 163 communicate with each other through the communication bus 164.

[0384] Memory 163, for storing computer programs;

[0385] The processor 161 is used to execute the program stored in the memory 163 to implement the following steps:

[0386] Obtain the original data in the table; obtain the initial processing results corresponding to the original data through the initial artificial intelligence model; in response to the modification operation on the initial processing result, obtain the modified initial processing result; wherein the modified initial processing result is used to provide the initial artificial intelligence model with learning the association between the original data and the modified initial processing result, so as to obtain the target artificial intelligence model.

[0387] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0388] The communication interface is used for communication between the above electronic device and other devices.

[0389] The memory may include a random access memory (RAM) or a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0390] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0391] In another embodiment provided in the present application, a storage medium is further provided. The storage medium stores instructions, which, when executed on a computer, enable the computer to execute the sample collection method described in any one of the above embodiments.

[0392] In another embodiment provided in the present application, a computer program product including instructions is also provided. When the computer program product is executed on a computer, the computer executes the sample collection method described in any one of the above embodiments.

[0393] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a storage medium, or transmitted from one storage medium to another storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0394] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are 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 other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0395] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0396] The above description is only a preferred embodiment of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A method for processing data in a table, characterized in that: The method comprises: Get the original data in the table; Obtaining an initial processing result corresponding to the original data through an initial artificial intelligence model; In response to a modification operation on the initial processing result, obtaining a modified initial processing result; Among them, the modified initial processing result is used to provide the initial artificial intelligence model with a learning relationship between the original data and the modified initial processing result to obtain a target artificial intelligence model.

2. The method according to claim 1, characterized in that: The obtaining of the original data in the table includes: Obtaining a form, in response to a confirmation operation of the AI ​​processing, displaying an AI settings page; The artificial intelligence setting page is used to set the data source, the processing result filling position and the label; In response to the setting operation of the data source, determining the original data in the set table; In response to a tag setting operation, determining the set tag; In response to a setting operation of a processing result filling position, determining a set processing result filling position; In response to a preset operation of the processing result, obtaining the original data in the table; The initial processing result corresponding to the original data is obtained by the initial artificial intelligence model, including: The initial artificial intelligence model is used to process the raw data according to the labels to obtain corresponding initial processing results.

3. The method according to claim 2, characterized in that The step of obtaining the original data in the table in response to the preset operation of the processing result includes: In response to the processing result preview operation, obtaining the original data in the table; The step of obtaining the modified initial processing result in response to the modification operation on the initial processing result includes: Displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page; In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

4. The method according to claim 2, characterized in that: The step of obtaining the original data in the table in response to the preset operation of the processing result includes: In response to the processing result generating operation, obtaining the original data in the table; The step of obtaining the modified initial processing result in response to the modification operation on the initial processing result includes: Fill the initial processing result corresponding to the original data in the processing result filling position; In response to a modification operation on the initial processing result, a modified initial processing result is obtained.

5. The method according to claim 1, characterized in that After obtaining the modified initial processing result in response to the modification operation on the initial processing result, the method further includes: Displaying first prompt information, where the first prompt information is used to prompt the initial artificial intelligence model to provide a processing sample; In response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction; In response to a first sample collection instruction, generating a first processed sample according to the original data and the modified initial processing result; The initial artificial intelligence model is trained based on the first processed sample to obtain a target artificial intelligence model.

6. The method according to claim 5, characterized in that After triggering a sample collection instruction in response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, the method further includes: In response to the second sample collection instruction, displaying a processing sample configuration page, the processing sample configuration page being used to configure a second processing sample; On the processing sample configuration page, obtaining target original data in a table input by a user and a target processing result corresponding to the target original data; generating a second processed sample according to the target original data and a target processing result corresponding to the target original data; The initial artificial intelligence model is trained based on the second processed sample to obtain a target artificial intelligence model.

7. The method according to claim 5, characterized in that Generating a first processed sample according to the original data and the modified initial processing result includes: Determining the modified initial processing result as the first expected processing result corresponding to the original data; A forward processing sample is generated according to the original data and the first expected processing result corresponding to the original data.

8. The method according to claim 5, characterized in that Generating a first processed sample according to the original data and the modified initial processing result includes: Determine the initial processing result before modification corresponding to the initial processing result after modification; Combining the preset reverse prompt information with the initial processing result before modification to generate a second expected processing result; A reverse processing sample is generated according to the original data and the second expected processing result.

9. The method according to claim 3, characterized in that: The step of displaying an artificial intelligence setting page in response to the confirmation operation of the artificial intelligence processing includes: In response to a selection operation of a cell in the table, displaying second prompt information, wherein the second prompt information is used to prompt that artificial intelligence will be used for processing, and there are multiple pairs of repeated labels in the cells in the column where the cell is located; In response to a confirmation operation of the artificial intelligence processing, an artificial intelligence setting page is displayed.

10. The method according to claim 9, characterized in that Before displaying the original data in the table and the initial processing results corresponding to the original data on the preview result page, the method further includes: Displaying the target original data in the table and the target processing results corresponding to the target original data on the preview result page; Wherein, the target original data is the original data in the table processed by the user, and the target processing result is the multiple pairs of repeated labels; In the preview result page, third prompt information is displayed, and the third prompt information is used to prompt the initial artificial intelligence model to provide a processing sample; In response to a confirmation operation of providing a processing sample to the initial artificial intelligence model, triggering a sample collection instruction; In response to a third sample collection instruction, acquiring the target original data and the target processing result corresponding to the target original data; Generate a third processed sample according to the target original data and the target processing result corresponding to the target original data; The initial artificial intelligence model is trained based on the third processed sample to obtain a target artificial intelligence model.

11. The method according to claim 2, further comprising: In response to the processing result generating operation, obtaining the original data in the table; Processing the original data according to the label through the target artificial intelligence model to obtain a corresponding new processing result; The new processing result corresponding to the original data is filled in the processing result filling position.

12. The method according to claim 1, characterized in that The method further comprises: In response to the instruction creation operation, displaying an instruction creation page, wherein the instruction creation page is used to set the persona, examples, and tasks in the instruction; In response to a persona setting operation, determining a set persona, wherein the persona is used to limit a working scope of an initial artificial intelligence model; In response to an example setting operation, determining a set example, wherein the example is used to instruct the initial artificial intelligence model to perform data processing with reference to the example; In response to the task setting operation, determining the set task, wherein the task is used to instruct the initial artificial intelligence model to process the target raw data in the table; In response to the processing result generation operation, providing the persona, examples, and tasks in the instruction to the initial artificial intelligence model; The processing result filling position of the table is filled with the processing result of the initial artificial intelligence model on the target original data in the table.

13. The method according to claim 12, characterized in that In response to the instruction creation operation, displaying an instruction creation page includes: In response to the selection operation of artificial intelligence processing, an artificial intelligence setting page is displayed, wherein the artificial intelligence setting page is used to set the data source, the processing result filling position and the label; In the artificial intelligence setting page, in response to a command center selection operation, a command center page is displayed; On the command center page, in response to a command creation operation, displaying a command creation page; The step of generating a response to the processing result, providing the persona, examples and tasks in the instruction to the initial artificial intelligence model, includes: In response to the command confirmation operation, displaying the command center page; On the command center page, in response to the confirmation operation, an artificial intelligence settings page is displayed; In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

14. The method according to claim 12, characterized in that In response to the instruction creation operation, displaying an instruction creation page includes: In response to the processing result preview operation, displaying the original data in the table and the initial processing result corresponding to the original data on the preview result page; In the preview result page, in response to the instruction viewing operation, displaying an instruction details page; On the instruction details page, in response to an instruction creation operation, displaying an instruction creation page; The generating operation of the response processing result, providing the persona, examples and tasks in the instruction to the large language model, comprises: In response to the command confirmation operation, displaying the command center page; On the command center page, in response to a confirmation operation, displaying the preview result page; In response to a page jump operation, jumping from the preview result page to an artificial intelligence setting page; In the artificial intelligence setting page, in response to a processing result generation operation, the persona, examples, and tasks in the instruction are provided to an initial artificial intelligence model.

15. The method according to claim 12, characterized in that The example includes table data, labels, and output locations, and in response to the example setting operation, determining the set example includes: In response to a setting operation of table data, determining example original data in the set table; In response to a label setting operation, determining a label corresponding to the set example original data; In response to the setting operation of the output position, a filling position corresponding to the set processing result of the example original data is determined.

16. The method according to claim 12, characterized in that The task includes table data, labels, and output locations, and in response to the task setting operation, determining the set task includes: In response to a setting operation of table data, determining target original data in the set table; In response to a tag setting operation, determining a tag corresponding to the set target original data; In response to the setting operation of the output position, the set filling position of the processing result of the target original data is determined.

17. A data processing device in a table, characterized in that: The device comprises: A data acquisition module is used to obtain the original data in the table; A data processing module, used to obtain an initial processing result corresponding to the original data through an initial artificial intelligence model; A result modification module, configured to obtain a modified initial processing result in response to a modification operation on the initial processing result; Among them, the modified initial processing result is used to provide the initial artificial intelligence model with a learning relationship between the original data and the modified initial processing result to obtain a target artificial intelligence model.

18. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, for implementing any of the methods described in claims 1-16 when executing a program stored in a memory.

19. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 16 is implemented.

Images