Data table processing method and device, electronic equipment and readable storage medium

Abstract

The invention provides a data table processing method and device, electronic equipment and a readable storage medium, and relates to the technical field of data analys.The data table processing method is applied to a terminal and comprises the steps that in response to a data configuration instruction for a target sequence in an initial data table, a configuration page is displayed; receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters; determining to-be-filled data corresponding to the target sequence according to the data processing strategy; and adding the to-be-filled data to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table. According to the invention, the user demand can be accurately and efficiently processed, so that the user satisfaction and the processing efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data analysis, and in particular, to a method, device, electronic device and readable storage medium for processing data tables. Background Art

[0002] With the rapid development of data analysis technology, there are more and more demands for processing various kinds of data, among which the demand for processing tabular data is becoming increasingly prominent. However, the data functions provided by current tables cannot accurately and efficiently process user requirements. Therefore, an effective solution is urgently needed to solve the above problems. Summary of the Invention

[0003] In view of the problems existing in the prior art, embodiments of the present invention provide a method, device, electronic device and readable storage medium for processing data tables.

[0004] The present invention provides a method for processing a data table, which is applied to a terminal and includes:

[0005] Responding to a data configuration instruction for a target sequence in an initial data table, and displaying a configuration page;

[0006] Receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters;

[0007] Determining to-be-filled data corresponding to the target sequence according to the data processing strategy;

[0008] Adding the to-be-filled data to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0009] According to a method for processing a data table provided by the present invention, the configuration parameters include a data source area, a processing type, and a processing target;

[0010] The determining a data processing strategy based on the configuration parameters includes:

[0011] Determining a data filling area according to the target sequence;

[0012] Determining a data processing strategy based on the processing type, the processing target, the data filling area, and the data source area.

[0013] According to a method for processing a data table provided by the present invention, the configuration parameters include a data processing strategy;

[0014] The determining a data processing strategy based on the configuration parameters includes:

[0015] Extract the data processing strategy from the configuration parameters, where the data processing strategy includes a data source area, a processing type, and a processing target.

[0016] According to a data table processing method provided by the present invention, determining the data to be filled corresponding to the target sequence according to the data processing strategy includes:

[0017] Based on the data processing strategy, obtain the data to be processed from the data source area;

[0018] Generate a data processing request according to the data to be processed, the processing type, and the processing target;

[0019] Send the data processing request to the server;

[0020] Receive the data to be filled feedback by the server based on the data processing request, where the data to be filled is obtained by the server processing the data to be processed based on the processing type and the processing target.

[0021] According to a data table processing method provided by the present invention, the data processing strategy is a calculation function or a macro;

[0022] The determining the data to be filled corresponding to the target sequence according to the data processing strategy includes:

[0023] In the case where the data processing strategy is a calculation function, determine the data to be filled corresponding to the target sequence according to the calculation function;

[0024] In the case where the data processing strategy is a macro, determine the data to be filled corresponding to the target sequence according to the macro.

[0025] According to a data table processing method provided by the present invention, the data processing strategy includes at least one of an information extraction strategy, a content summary strategy, a content classification strategy, an emotion recognition strategy, a content translation strategy, and a custom processing strategy;

[0026] The information extraction strategy is used to extract key information in the data source area;

[0027] The content summary strategy is used to summarize the content in the data source area;

[0028] The content classification strategy is used to classify the content in the data source area;

[0029] The emotion recognition strategy is used to recognize the emotion expressed by the content in the data source area;

[0030] The content translation strategy is used to translate the content in the data source area;

[0031] The custom processing strategy is used to perform custom processing on the content in the data source area, and the custom processing is determined based on user requirements.

[0032] According to a data table processing method provided by the present invention, the information extraction strategy includes a specified information extraction strategy, a multimodal information extraction strategy, and a cross-data source information extraction strategy;

[0033] The specified information extraction strategy is used to extract specified information in the data source area, and the specified information is preset or specified by the user;

[0034] The multimodal information extraction strategy is used to extract information from multimodal content in the data source area, and the multimodal content includes at least one of pictures, attachments, and links;

[0035] The cross-data source information extraction strategy is used to extract specific content from other data sources, and the semantic similarity between the specific content and the selected content is greater than the similarity threshold. The other data sources include other data tables other than the initial data table or other files other than the specified file, and the specified file is the file corresponding to the initial data table.

[0036] According to a data table processing method provided by the present invention, receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters, includes:

[0037] Receiving at least two configuration parameters based on the configuration page;

[0038] Determining sub-data processing strategies respectively corresponding to the configuration parameters;

[0039] Fusing the sub-data processing strategies to obtain a data processing strategy.

[0040] According to a data table processing method provided by the present invention, the method further includes:

[0041] Receiving a selection instruction for data to be processed in the initial data table, and displaying a data processing page, where the data processing page includes at least one data processing intention, and the data processing intention is used to jump to the configuration page corresponding to the data processing intention.

[0042] The present invention also provides a data table processing method, applied to a server, including:

[0043] Receive a data processing request sent by a receiving terminal. The data processing request is generated based on data to be processed, a processing type, and a processing target. The data to be processed, the processing type, and the processing target are determined based on a data processing strategy. The data processing strategy is obtained from a configuration page for display based on a data configuration instruction for a target sequence in an initial data table.

[0044] Process the data to be processed according to the processing type and the processing target to obtain data to be filled.

[0045] Send the data to be filled to the terminal so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0046] According to a data table processing method provided by the present invention, the step of processing the data to be processed according to the processing type and the processing target to obtain data to be filled includes:

[0047] Determine model input parameters according to the processing type, the processing target, and the data to be processed.

[0048] Input the model input parameters into an artificial intelligence model for processing to obtain data to be filled.

[0049] According to a data table processing method provided by the present invention, the model input parameters include a prompt statement, an example, a guiding statement, and sampling parameters. The sampling parameters include an output token parameter and a temperature parameter.

[0050] According to a data table processing method provided by the present invention, the artificial intelligence model includes a conversion processing layer, a probability prediction layer, and a sampling layer.

[0051] The step of inputting the model input parameters into an artificial intelligence model for processing to obtain data to be filled includes:

[0052] Convert the prompt statement, the example, and the guiding statement into processing units through the conversion processing layer, and perform data processing on the processing units to obtain data to be output.

[0053] Predict the probability of the data to be output based on the temperature parameter through the probability prediction layer.

[0054] Sample the data to be output based on the output token parameter and the probability through the sampling layer to obtain data to be filled.

[0055] The present invention also provides a data table processing device applied to a terminal, including:

[0056] A display module, configured to display a configuration page in response to a data configuration instruction for a target sequence in an initial data table;

[0057] A first determination module, configured to receive configuration parameters based on the configuration page and determine a data processing strategy based on the configuration parameters;

[0058] A second determination module, configured to determine data to be filled corresponding to the target sequence according to the data processing strategy;

[0059] An addition module, configured to add the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0060] The present invention further provides a data table processing device, which is applied to a server and includes:

[0061] A first receiving module, configured to receive a data processing request sent by a terminal, where the data processing request is generated based on data to be processed, a processing type, and a processing target, the data to be processed, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from a configuration page displayed in response to a data configuration instruction for a target sequence in an initial data table;

[0062] A processing module, configured to process the data to be processed according to the processing type and the processing target to obtain data to be filled;

[0063] A sending module, configured to send the data to be filled to the terminal so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0064] The present invention further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where when the processor executes the program, the data table processing method described in any one of the above is implemented.

[0065] The present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the data table processing method described in any one of the above is implemented.

[0066] The present invention further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the data table processing method described in any one of the above is implemented.

[0067] The data table processing method, device, electronic device and readable storage medium provided by the present invention display a configuration page by responding to a data configuration instruction for a target sequence in an initial data table; receive configuration parameters based on the configuration page, and determine a data processing strategy based on the configuration parameters; determine to-be-filled data corresponding to the target sequence according to the data processing strategy; and add the to-be-filled data to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table. By generating a data processing strategy through configuration parameters and then obtaining the to-be-filled data, the present invention can accurately and efficiently process user requirements, thereby improving user satisfaction and processing efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0069] Figure 1 is one of the flow diagrams of the data table processing method provided by the present invention;

[0070] Figure 2 is one of the interface diagrams of the data table processing method provided by the present invention;

[0071] Figure 3 is the second of the interface diagrams of the data table processing method provided by the present invention;

[0072] Figure 4 is the third of the interface diagrams of the data table processing method provided by the present invention;

[0073] Figure 5 is the fourth of the interface diagrams of the data table processing method provided by the present invention;

[0074] Figure 6 is the fifth of the interface diagrams of the data table processing method provided by the present invention;

[0075] Figure 7 is the second of the flow diagrams of the data table processing method provided by the present invention;

[0076] Figure 8 is the sixth of the interface diagrams of the data table processing method provided by the present invention;

[0077] Figure 9 is the seventh of the interface diagrams of the data table processing method provided by the present invention;

[0078] Figure 10It is the eighth schematic diagram of the interface of the data table processing method provided by the present invention;

[0079] Figure 11 It is the ninth schematic diagram of the interface of the data table processing method provided by the present invention;

[0080] Figure 12 It is the third schematic diagram of the flow of the data table processing method provided by the present invention;

[0081] Figure 13 It is the fourth schematic diagram of the flow of the data table processing method provided by the present invention;

[0082] Figure 14 It is one of the schematic diagrams of the structure of the data table processing device provided by the present invention;

[0083] Figure 15 It is the second schematic diagram of the structure of the data table processing device provided by the present invention;

[0084] Figure 16 It is the schematic diagram of the structure of the electronic device provided by the present invention. Detailed implementation manners

[0085] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0086] To facilitate a clearer understanding of the embodiments of the present invention, some relevant background knowledge is introduced as follows.

[0087] In a specific scenario of processing tabular data, there is the following data:

[0088] 1. User feedback requires marking and analyzing positive and negative directions;

[0089] 2. Address information requires extracting information such as province and city according to requirements;

[0090] 3. Product data in cross-border e-commerce business needs to be translated into different languages;

[0091] 4. A large amount of evaluation text in interview information needs to be condensed into a short talent profile.

[0092] Currently, the functions in the table, such as simple formulas, the functions corresponding to Ctrl+E, etc., cannot accurately and efficiently process.

[0093] The following is combined withFigures 1 - 16 Describe a data table processing method, apparatus, electronic device, and readable storage medium of the present invention.

[0094] Figure 1 It is one of the flow schematic diagrams of the data table processing method provided by the present invention. Refer to Figure 1 As shown, it is applied to a terminal and includes step 101-step 104, where:

[0095] Step 101: In response to a data configuration instruction for a target sequence in the initial data table, display a configuration page.

[0096] First of all, it should be noted that the execution subject of the present invention can be any electronic device that processes data tables, such as any one of a smart phone, a smart watch, a desktop computer, a laptop computer, etc.

[0097] Specifically, the initial data table refers to the data table before data processing. The data table can be a table in a table file, a table in a text file, a table in a presentation file, etc. The target sequence refers to a certain one or a certain column in the initial data table, which can be the sequence of data sources or the sequence used to fill data. The data configuration instruction refers to an instruction for coordinating the data corresponding to the target sequence. The configuration page is used for data configuration.

[0098] In practical applications, the terminal can display the initial data table through the front-end interface. After the user selects the target sequence in the initial data table, the user can trigger the data configuration instruction through an operation. Correspondingly, the terminal receives the data configuration instruction and displays the configuration page to the user through the front-end interface for the user to perform data configuration.

[0099] Exemplarily, refer to Figure 2 , Figure 2 It is one of the interface schematic diagrams of the data table processing method provided by the present invention: In the initial data table, the user selects the target sequence (the E column, the column of user emotion analysis), and clicks the right mouse button to trigger the configuration option interface. The user selects AI automatic filling in the configuration option interface, that is, triggers the data configuration instruction.

[0100] Step 102: Receive configuration parameters based on the configuration page, and determine a data processing strategy based on the configuration parameters.

[0101] Specifically, the configuration parameters are used to characterize the content of data configuration. The data processing strategy refers to the way and method of data processing, etc.

[0102] In practical applications, the user performs data configuration in the configuration page, that is, fills in the configuration parameters. Correspondingly, the terminal receives the configuration parameters filled in by the user based on the configuration page. Further, the terminal analyzes the configuration parameters to obtain a data processing strategy.

[0103] Step 103: Determine the data to be filled corresponding to the target sequence according to the data processing strategy.

[0104] Specifically, the data to be filled refers to the generated data.

[0105] In practical applications, on the basis of determining the data processing strategy, further, based on the data processing strategy, obtain the corresponding data to be processed, and process the data to be processed according to the data processing strategy, so as to obtain the data to be filled.

[0106] Step 104: Add the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0107] Specifically, the data filling area refers to the area for filling data. When the target sequence is the sequence of the data source, the data filling area corresponding to the target sequence is the row or column where the generated sequence corresponding to the target sequence is located; when the target sequence is the sequence of the data to be filled, the data filling area corresponding to the target sequence is the row or column where the target sequence is located.

[0108] In practical applications, when the data to be filled is obtained, fill the data to be filled in the data filling area corresponding to the target sequence in the initial data table, so as to obtain the data table after data processing, that is, the target data table.

[0109] Further, the target data table can also be displayed to the user through the front-end interface for the user to view the data processing result or perform processing such as adjustment and correction on the data processing result.

[0110] Exemplarily, on the basis of Figure 2 refer to Figure 3 , Figure 3 This is the second schematic diagram of the interface of the data table processing method provided by the present invention: after the data to be filled is obtained, fill the data to be filled in the E column, that is, the column of "user emotion analysis", to obtain the target data table.

[0111] The data table processing method provided by the present invention displays a configuration page by responding to a data configuration instruction for a target sequence in an initial data table; receives configuration parameters based on the configuration page, and determines a data processing strategy based on the configuration parameters; determines the data to be filled corresponding to the target sequence according to the data processing strategy; adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table. The present invention generates a data processing strategy through configuration parameters, and then obtains the data to be filled, which can accurately and efficiently process user requirements, and then improve user satisfaction and processing efficiency.

[0112] In one or more alternative embodiments of the present invention, the configuration parameters include a data source area, a processing type, and a processing target; based on the configuration parameters, a data processing strategy is determined, and the specific implementation process may be as follows:

[0113] Determine a data filling area according to the target sequence;

[0114] Based on the processing type, the processing target, the data filling area, and the data source area, determine a data processing strategy.

[0115] Specifically, the data source area represents the position of the data to be processed in the initial data table. The processing type refers to the type of processing to be performed on the data to be processed, and the processing type may include at least one of information extraction, content summarization, content classification, sentiment recognition, content translation, and custom processing, etc. The processing target represents the target of data processing for the data to be processed.

[0116] Exemplarily, in the case where the processing type is information extraction, the processing target may be to extract address information, identity information, etc.; in the case where the processing type is content summarization, the processing target may be monthly summary, annual summary, sales summary, personal experience summary, etc.; in the case where the processing type is content classification, the processing target may be to be divided into N categories, namely the first category to the Nth category, where N is a positive integer greater than 1, such as being divided into four categories: spring, summer, autumn, and winter; in the case where the processing type is sentiment recognition, the processing target may be positive, neutral, negative, preferred, disgusted, etc.; in the case where the processing type is content translation, the processing target may be to translate into English, translate into Korean, translate into French, translate into Chinese, etc.; in the case where the processing type is custom processing, the processing target may be a user-defined target, such as translating into English and extracting address information.

[0117] In practical applications, the area that needs to be filled with data, that is, the data filling area, can be determined based on the target sequence. And according to the processing type, the processing target, the data filling area, and the data source area, a data processing strategy is determined, that is, which data to process, what kind of processing to perform, what to process into, and where to fill it.

[0118] Exemplarily, on the basis of Figure 2 refer to Figure 4 , Figure 4This is the third schematic diagram of the interface of the data table processing method provided by the present invention: upon receiving a data configuration instruction, a configuration page is displayed, that is, list type configuration. Subsequently, the user can select AI capabilities, that is, processing types, including intelligent extraction (information extraction), intelligent classification (content), content summary, sentiment analysis (sentiment recognition), and custom tasks (custom processing). Here, intelligent classification is taken as an example; a data source column can also be added, that is, the data source area. Here, "feedback (column D)" is taken as an example; the desired classification can also be described, that is, the processing target. Here, taking the classification into three categories: neutral, positive, and negative as an example. Correspondingly, the configuration parameters include the data source area, processing type, and processing target.

[0119] In this embodiment, the terminal can generate a data processing strategy based on the data source area, processing type, and processing target configured by the user. In this way, the accuracy and speed of determining the data processing strategy can be improved.

[0120] In one or more alternative embodiments of the present invention, the configuration parameters include a data processing strategy; based on the configuration parameters, determining the data processing strategy, the specific implementation process can be as follows:

[0121] Extract the data processing strategy from the configuration parameters, and the data processing strategy includes a data source area, a processing type, and a processing target.

[0122] Specifically, the data source area represents the position of the data to be processed in the initial data table. The processing type refers to the type of processing to be performed on the data to be processed, and the processing type can include at least one of information extraction, content summary, content classification, emotion recognition, content translation, and custom processing, etc. The processing target represents the target of data processing for the data to be processed.

[0123] In practical applications, the user can directly configure the data processing strategy. On this basis, the terminal can directly use the data processing strategy configured by the user. In this way, the speed of determining the data processing strategy can be improved.

[0124] Exemplarily, on the basis of Figure 2 refer to Figure 5 Figure 5 This is the fourth schematic diagram of the interface of the data table processing method provided by the present invention: upon receiving a data configuration instruction, a configuration page is displayed, and the configuration page contains multiple data processing strategies. Here, the data processing strategy is taken as a formula as an example. When the user selects a certain data processing strategy among them, correspondingly, the configuration parameters include the data processing strategy.

[0125] Exemplarily, on the basis of Figure 2 refer to Figure 6 Figure 6Figure 5 is a schematic diagram of the interface of the data table processing method provided by the present invention: upon receiving a data configuration instruction, a configuration page is displayed. The configuration page is an input box, and the user can input a data processing strategy in the input box. Here, the data processing strategy is taken as an example of a formula. Correspondingly, the configuration parameters include the data processing strategy.

[0126] In an implementable embodiment of the present invention, the process of determining the data to be filled corresponding to the target sequence according to the data processing strategy may be specifically implemented as follows:

[0127] Obtain the data to be processed from the data source area based on the data processing strategy;

[0128] Generate a data processing request according to the data to be processed, the processing type, and the processing target;

[0129] Send the data processing request to the server;

[0130] Receive the data to be filled feedback by the server based on the data processing request. The data to be filled is obtained by the server processing the data to be processed based on the processing type and the processing target.

[0131] Specifically, the data to be processed refers to the data that needs to be processed, that is, the source data. The data processing request refers to the data processing request sent to the server.

[0132] In practical applications, since the data processing ability of the server is relatively stronger than that of the terminal, the server can be used to determine the data to be filled.

[0133] First, based on the data processing strategy, obtain the data to be processed from the data source area. The data to be processed includes but is not limited to structured data, unstructured data, semi-structured data, etc.

[0134] Then, generate a data processing request according to the data to be processed, the processing type, and the processing target, that is, generate a data processing request carrying the data to be processed, the processing type, and the processing target. The data processing request includes but is not limited to data cleaning, data conversion, data mining, etc.

[0135] Furthermore, the terminal sends the data processing request to the server. The server can be one or more, which can be a distributed data processing cluster, or an independent server or a cloud computing platform.

[0136] After receiving the data processing request, the server can process the data to be processed based on the processing type and the processing target in the data processing request to obtain the data to be filled, and then feedback the data to be filled to the terminal. Correspondingly, the terminal receives the data to be filled.

[0137] Exemplarily, the server determines model input parameters according to the processing type, processing target, and processing data, where the model input parameters include a prompt statement, examples, guiding statements, and sampling parameters, and the sampling parameters include output token parameters and / or temperature parameters; then the model input parameters are input into a large language model for processing to obtain data to be filled.

[0138] See Figure 7 , Figure 7 is the second flowchart of the data table processing method provided by the present invention:

[0139] Step 701: The terminal displays a configuration page in response to a data configuration instruction for a target sequence in the initial data table.

[0140] Step 702: The terminal receives configuration parameters based on the configuration page and determines a data processing strategy based on the configuration parameters.

[0141] Step 703: The terminal obtains data to be processed from the data source area based on the data processing strategy.

[0142] Step 704: The terminal generates a data processing request according to the data to be processed, processing type, and processing target.

[0143] Step 705: The terminal sends the data processing request to the server.

[0144] Step 706: The server processes the data to be processed according to the processing type and processing target in the data processing request to obtain data to be filled.

[0145] Step 707: The server sends the data to be filled to the terminal.

[0146] Step 708: The terminal adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0147] In this way, by determining the data to be filled by the server, the accuracy and efficiency of the data to be filled can be improved. In addition, by combining data processing with a large model, the universality and applicability of data processing can be improved.

[0148] In another implementable embodiment of the present invention, the specific implementation process of determining the data to be filled corresponding to the target sequence according to the data processing strategy may be as follows:

[0149] Obtain data to be processed from the data source area based on the data processing strategy;

[0150] Determine model input parameters according to the processing type, the processing target, and the data to be processed;

[0151] Input the model input parameters into a large language model for processing to obtain the data to be filled.

[0152] In practical applications, the terminal can independently determine the data to be filled based on a large language model. First, based on a data processing strategy, obtain the data to be processed from the data source area, and the data to be processed includes but is not limited to structured data, unstructured data, semi-structured data, etc. Then, determine the model input parameters according to the processing type, processing target, and processing data, where the model input parameters include a prompt statement, examples, guiding statements, and sampling parameters, and the sampling parameters include output token parameters and / or temperature parameters; then input the model input parameters into a large language model for processing to obtain the data to be filled.

[0153] In this way, the terminal independently determines the data to be filled, which can not only ensure the security and privacy of the data in the data table, but also avoid the phenomenon that when there is no network, it is impossible to connect to the network and thus impossible to obtain the data to be filled, that is, to a certain extent, improve the success rate of determining the data to be filled. In addition, by combining the data processing with the large model, the generality and applicability of data processing can be improved.

[0154] In one or more alternative embodiments of the present invention, the data processing strategy is a calculation function or a macro; the process of determining the data to be filled corresponding to the target sequence according to the data processing strategy may be specifically implemented as follows:

[0155] In the case where the data processing strategy is a calculation function, determine the data to be filled corresponding to the target sequence according to the calculation function;

[0156] In the case where the data processing strategy is a macro, determine the data to be filled corresponding to the target sequence according to the macro.

[0157] Specifically, the calculation function here can be a common formula, such as a summation formula, an average formula, etc., or an artificial intelligence (AI) formula, such as an information extraction formula, a content summary formula, a content classification formula, an emotion recognition formula, a content translation formula, and a custom processing formula, etc., which are used to represent some formulas with tabular data as input and text content as output. A macro refers to replacing a certain text pattern according to a series of predefined rules, such as code.

[0158] In practical applications, when processing a data table, a calculation function can be used as the data processing strategy to determine the data to be filled, or a macro can be used as the data processing strategy to determine the data processing strategy. In this way, the generality and applicability of data table processing can be improved.

[0159] In one or more alternative embodiments of the present invention, the data processing strategy includes at least one of an information extraction strategy, a content summarization strategy, a content classification strategy, an emotion recognition strategy, a content translation strategy, and a custom processing strategy;

[0160] The information extraction strategy is used to extract key information from the data source area;

[0161] The content summarization strategy is used to summarize the content in the data source area;

[0162] The content classification strategy is used to classify the content in the data source area;

[0163] The emotion recognition strategy is used to recognize the emotion expressed by the content in the data source area;

[0164] The content translation strategy is used to translate the content in the data source area;

[0165] The custom processing strategy is used to perform custom processing on the content in the data source area, and the custom processing is determined based on user requirements.

[0166] Specifically, the data table processing mainly supports the following functions:

[0167] 1) Extract the information of the cell, that is, the information extraction function; the information extraction function corresponds to the information extraction strategy, and the information extraction strategy is used to extract the data source area, that is, the key information contained in the cell in the table. Among them, the key information can be user requirement information, such as specified information (such as address information, identity information, etc.), multimodal information, and cross-data source information, etc.

[0168] 2) Summarize the content of the target cell, which is used to summarize large paragraphs of text content, that is, the content summarization function, such as summarizing a weekly report, etc.; the content summarization function corresponds to the content summarization strategy, and the content summarization strategy is used to summarize the content in the data source area (target cell).

[0169] 3) Classify the content of the target cell, that is, the content classification function, such as classifying different occupations into the education industry, the construction industry, etc.; the content classification function corresponds to the content classification strategy, and the content classification strategy is used to classify the content in the data source area (target cell).

[0170] 4) Classify the text into different emotions, which is used to classify user feedback into NPS (Net Promoter Score) or sentiment, namely, the emotion recognition function; the emotion recognition function corresponds to the emotion recognition strategy, which is used to identify the emotions expressed by the content in the data source area, that is, to classify the content in the data source area into different emotions.

[0171] 5) Translate the content of the target cell, that is, the content translation function; the content translation function corresponds to the content translation strategy, and the content translation strategy is used to translate the content in the data source area.

[0172] 6) Users ask customized questions about cell contents, and the artificial intelligence model answers, which can cover all the above scenarios and has a higher degree of freedom, namely, the custom processing function; the custom processing function corresponds to the custom processing strategy, and the custom processing strategy is used to perform custom processing on the content in the data source area, and the custom processing is determined based on user needs.

[0173] It should be noted that the customized processing may be at least one of information extraction, content summarization, content classification, emotion recognition, content translation and data analysis on the content in the data source area.

[0174] Exemplarily, the custom processing may be to first translate the content in the data source area, and then extract information from the translated content. For example, if the content in the data source area is "My home address is No. E, DStreet, C Village, B City, A Province", the content is translated (translated into Chinese) to obtain the translated content "My home address is No. E, D Street, C Village, B City, A Province", and then information extraction (extracting address information) is performed to obtain "No. E, D Street, C Village, B City, A Province".

[0175] In this way, not only can the application scope of data table processing be expanded, but also a variety of choices can be provided to users, thereby improving user satisfaction.

[0176] In one or more optional embodiments of the present invention, the information extraction strategy includes a specified information extraction strategy, a multimodal information extraction strategy, and a cross-data source information extraction strategy;

[0177] The specified information extraction strategy is used to extract the specified information in the data source area, and the specified information is preset or specified by the user;

[0178] The multimodal information extraction strategy is used to extract information of multimodal content in the data source area, wherein the multimodal content includes at least one of a picture, an attachment, and a link;

[0179] The cross-data-source information extraction strategy is used to extract specific content from other data sources. The semantic similarity between the specific content and the selected content is greater than the similarity threshold. The other data sources include other data tables outside the initial data table or other files outside the specified file, and the specified file is the file corresponding to the initial data table.

[0180] In practical applications, the specified information extraction strategy can be used to extract specified information from cells, mainly for extracting data that is difficult to process by traditional table functions such as specified provinces, cities, districts, and identity information. That is, the specified information can be address information, identity information, etc.

[0181] The multi-modal information extraction strategy is used for multi-modal information, which can be at least one of the information in pictures, attachments, and links, that is, the information in multi-modal content. That is, the multi-modal information extraction strategy extracts the information in pictures, attachments, and links in the table and processes it according to the user's requirements, such as extracting key information.

[0182] The cross-data-source information extraction strategy is used to find semantically similar data across tables and files, that is, content.

[0183] It should be noted that the cross-data-source information extraction strategy can combine the Embedding ability of the large language model to find semantically similar data across tables and files, rather than data with exactly the same characters. There are mainly the following two situations:

[0184] 1) The Embedding ability can convert all existing large amounts of data and the user's input into vectors for calculation, which can help users find all semantically similar data while saving consumption.

[0185] Exemplarily, see Figure 8 , Figure 8 is the sixth interface schematic diagram of the data table processing method provided by the present invention: When crossing tables, a selected area can be specified. Here, "[C2:C19]" is taken as an example. Then, the corresponding attachment can be selected through the text selection control (the input box corresponding to the attachment content extraction), and some data on the cross-table or cross-file can be selected through the data selection control (the input box corresponding to the extraction of all data) for data search.

[0186] 2) Through the Embedding ability, text can be transformed into a high-dimensional vector representation to capture the semantic relationships between words and sentences, thereby enabling tasks such as semantic similarity calculation, text classification, and information retrieval. Compared with directly inputting the text to the model for answering, before that, relevant contexts can be retrieved from the specified knowledge base through the Embedding function and put into the model's input, allowing the model to make a smoother and more reasonable answer based on the context. Based on this, capabilities such as long-term memory retrieval and knowledge base retrieval can be achieved.

[0187] Exemplarily, refer to Figure 9 , Figure 9 which is the seventh schematic diagram of the interface of the data table processing method provided by the present invention: When a region is selected in the initial data table, a prompt box appears, which displays the selected region. Here, "[A5:B5]" is taken as an example. Then, the corresponding selection function is selected through the function selection control (input box for approximate data retrieval), and the corresponding file or worksheet is selected through the file or worksheet selection control (input box corresponding to the current file, current worksheet, another file, etc.) for data retrieval, etc.

[0188] Exemplarily, on the basis of Figure 9 , refer to Figure 10 , Figure 10 which is the eighth schematic diagram of the interface of the data table processing method provided by the present invention: Here, the selection of the current file "Student Information Table" is used for illustration.

[0189] In one or more alternative embodiments of the present invention, the configuration parameters are received based on the configuration page, and based on the configuration parameters, a data processing strategy is determined. The specific implementation process can be as follows:

[0190] Receive at least two configuration parameters based on the configuration page;

[0191] Determine the sub-data processing strategies respectively corresponding to the configuration parameters;

[0192] Fuse the sub-data processing strategies to obtain a data processing strategy.

[0193] In practical applications, considering that the present invention is a content generation method combining strategies, certain capabilities can be achieved through combined strategies.

[0194] Specifically, a set of configuration parameters can be received through the configuration page, and then the user is guided to perform data configuration again on the configuration page. Correspondingly, the terminal receives at least two sets of configuration parameters.

[0195] Further, each group of configuration parameters can be fused to obtain fused configuration parameters, and then a data processing strategy can be obtained based on the fused configuration parameters. Alternatively, sub-data processing strategies corresponding to each group of configuration parameters can be determined first, and then the sub-data processing strategies can be fused to obtain a data processing strategy.

[0196] In this way, the generality and applicability of data table processing can be further improved, providing users with more data table processing methods, and thus improving user satisfaction.

[0197] In one or more alternative embodiments of the present invention, the method further includes:

[0198] Receiving a selection instruction for the data to be processed in the initial data table, and presenting a data processing page, where the data processing page includes at least one data processing intention, and the data processing intention is used to jump to a configuration page corresponding to the data processing intention.

[0199] Specifically, considering that the present invention can be an interaction for batch-generated content: content can be batch-generated in the data table. That is, when selecting the data to be processed, an interactive window such as an in-text floating window and a side task pane is provided for the user, that is, the data processing page. The interface exposed to the user may have different interaction forms. In this way, user satisfaction can be improved.

[0200] Exemplarily, refer to Figure 11 , Figure 11 is the ninth schematic diagram of the interface of the data table processing method provided by the present invention: When a region is selected in the initial data table, a prompt box appears, which shows the selected region. Here, "[C2:C19]" is taken as an example. Then, the corresponding selection function is selected through the function selection control (data extraction, intelligent analysis, approximate data retrieval, formula generation, attachment content extraction, data analysis, input box corresponding to the instruction center), and the content to be extracted is input through the content output box; for example, address, company, etc.

[0201] Figure 12 is the third schematic diagram of the flow of the data table processing method provided by the present invention. Refer to Figure 12 shown, applied to a server, including step 1201 - step 1204, where:

[0202] Step 1201: Receiving a data processing request sent by a terminal, where the data processing request is generated based on the data to be processed, the processing type, and the processing target, and the data to be processed, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from a configuration page presented based on a data configuration instruction for a target sequence in the initial data table.

[0203] In practical applications, the terminal can display an initial data table through the front-end interface. After the user selects a target sequence in the initial data table, a data configuration instruction can be triggered through an operation. Correspondingly, the terminal receives the data configuration instruction and displays a configuration page to the user through the front-end interface for the user to perform data configuration.

[0204] The user performs data configuration in the configuration page, that is, fills in configuration parameters. Correspondingly, the terminal receives the configuration parameters filled in by the user based on the configuration page. Further, the terminal analyzes the configuration parameters to obtain a data processing strategy.

[0205] Since the data processing ability of the server is relatively more powerful than that of the terminal, the server can be used to determine the data to be filled: based on the data processing strategy, obtain the data to be processed from the data source area, and the data to be processed includes but is not limited to structured data, unstructured data, semi-structured data, etc.; then, generate a data processing request according to the data to be processed, the processing type, and the processing target, that is, generate a data processing request carrying the data to be processed, the processing type, and the processing target, and the data processing request includes but is not limited to data cleaning, data conversion, data mining, etc.; further, the terminal sends the data processing request to the server, where the server can be one or more, and can be a distributed data processing cluster, or an independent server or a cloud computing platform.

[0206] Step 1202: Process the data to be processed according to the processing type and the processing target to obtain the data to be filled.

[0207] In practical applications, after receiving the data processing request, the server can process the data to be processed based on the processing type and the processing target in the data processing request to obtain the data to be filled.

[0208] Step 1203: Send the data to be filled to the terminal so that the terminal adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0209] In practical applications, after obtaining the data to be filled, the server feeds back the data to be filled to the terminal. Correspondingly, the terminal receives the data to be filled.

[0210] In the case of obtaining the data to be filled, the terminal fills the data to be filled in the data filling area corresponding to the target sequence in the initial data table, thereby obtaining a data table after data processing, that is, a target data table.

[0211] The data table processing method provided by the present invention receives a data processing request sent by a terminal. The data processing request is generated based on data to be processed, a processing type, and a processing target. The data to be processed, the processing type, and the processing target are determined based on a data processing strategy. The data processing strategy is obtained from a configuration page for display based on a data configuration instruction for a target sequence in an initial data table. According to the processing type and the processing target, the data to be processed is processed to obtain data to be filled. The data to be filled is sent to the terminal so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table. By generating a data processing strategy through configuration parameters and then obtaining the data to be filled, the present invention can accurately and efficiently process user requirements, thereby improving user satisfaction and processing efficiency.

[0212] In one or more alternative embodiments of the present invention, the process of processing the data to be processed according to the processing type and the processing target to obtain the data to be filled may be specifically implemented as follows:

[0213] Determine model input parameters according to the processing type, the processing target, and the data to be processed;

[0214] Input the model input parameters into an artificial intelligence model for processing to obtain the data to be filled.

[0215] Specifically, the model input parameters include a prompt, a sample message, a glyph, and sampling parameters. The sampling parameters include an output token parameter (top_p) and a temperature parameter (temperature). The artificial intelligence model may be a deep neural network model, a large language model, or any other artificial intelligence model that can determine the data to be filled.

[0216] In practical applications, the server splices parameters such as prompt, sample_message, glyph, top_p, and temperature according to the processing type, the processing target, and the data to be processed, and requests the artificial intelligence model. Further, the artificial intelligence model performs processing such as data generation or prediction based on parameters such as prompt, sample_message, glyph, top_p, and temperature to obtain the data to be filled. In this way, determining the data to be filled model through the artificial intelligence model can improve the accuracy of the data to be filled.

[0217] In one or more alternative embodiments of the present invention, the artificial intelligence model includes a conversion processing layer, a probability prediction layer, and a sampling layer; accordingly, the process of inputting the model input parameters into the artificial intelligence model for processing to obtain the data to be filled can be specifically implemented as follows:

[0218] Through the conversion processing layer, convert the prompt statement, the example, and the guiding statement into processing units, and perform data processing on the processing units to obtain the data to be output;

[0219] Through the probability prediction layer, predict the probability of the data to be output based on the temperature parameter;

[0220] Through the sampling layer, sample the data to be output based on the output token parameter and the probability to obtain the data to be filled.

[0221] Specifically, the processing unit can be a processing unit for the artificial intelligence model to perform text or data processing, that is, a token. A token can be a word, punctuation mark, number, or other language element.

[0222] It should be noted that after the artificial intelligence model obtains the request parameters from the server, (the conversion processing layer) will convert parameters such as prompt, sample_message, and glyph into tokens for processing to obtain the data to be output. Specifically, the artificial intelligence model divides each language element in prompt, sample_message, and glyph, such as words, punctuation marks, numbers, or other language elements, to obtain the tokens corresponding to prompt, sample_message, and glyph respectively; and guided by the tokens of sample_message and the tokens of glyph, perform data recognition and processing index determination on the tokens of prompt to obtain the tokens of the data to be processed and the processing index, and process the tokens of the data to be processed according to the processing index to obtain the processed tokens, that is, the data to be output.

[0223] Moreover, (the probability prediction layer) predicts the probability of the text (processed tokens) to be output according to the configuration of temperature, that is, predicts the probability of each processed token being output.

[0224] Finally (the sampling layer), according to the configuration of top_p and the probability, extract the target tokens with high probability and belonging to top_p from each processed token, that is, sample the data to be output, then splice the target tokens to obtain the finally output result, that is, the data to be filled, and return the data to be filled to the server.

[0225] In this way, the accuracy of the data to be filled can be improved.

[0226] The following further describes the data table processing method provided by the present invention in conjunction with Figure 13 the following. Figure 13 For the fourth flow diagram of the data table processing method provided by the present invention, see Figure 13 As shown, a front-end entry for batch processing data in a whole column is provided for the text data processing problems that users often encounter in their work:

[0227] A user has a batch of data to be classified. Open the front-end interface for configuration classification, that is, configure through the front-end interface of the terminal (table kernel) (AI), and then perform configuration (including 1. classification categories and 2. data sources).

[0228] Correspondingly, the front-end interface sends "1. classification categories, 2. data sources, and 3. data generation locations" to the terminal. The terminal will automatically set up the AI formula in the corresponding cells according to the user's configuration, and send a request to the server (table server side) (including 1. data to be processed and 2. categories to be classified) to obtain the calculation result.

[0229] After the server receives the request from the terminal, it will splice parameters such as prompt, sample_message, glyph, top_p, and temperature according to the configuration informed by the terminal, and request the large language model, that is, splice the information given by the kernel into the prompt, and transmit 1. prompt, 2. sample_message (example, determined according to the user's configuration), 3. glyph (constraint, determined according to the user's configuration), and 4. top_p / temperature (determining the sampling probability of the generated result) to the large language model.

[0230] After the large language model receives the request parameters from the server, it will convert parameters such as prompt, sample_message, and glyph into tokens for processing, that is, process prompt / sample_message / glyph; and according to the configuration of temperature, predict the probability of the text to be output, that is, generate a candidate queue of the prediction result and its probability according to the configuration of temperature; finally, sample according to the configuration of top_p to obtain the final output result, that is, determine what the candidate word for the final sampling is according to the configuration of top_p, and send the processed result to the server.

[0231] The server receives the results from the large language model, performs post-processing, that is, processes them into results that can be displayed, and provides these results to the kernel. The kernel then returns this result to the table front end, and finally the table front end performs rendering and displays it on the screen, that is, presents the calculation result to the user.

[0232] In addition to the user using the front-end interface for processing, the user can also directly use AI functions for processing, and the degree of freedom of processing will not be limited to the entire column. The following functions are mainly supported:

[0233] 1) Extract the specified information of the cell, mainly used to extract data such as the specified province, city, district, and ID card that are difficult to process by traditional table functions;

[0234] 2) Summarize the content of the target cell, used to summarize large sections of text, such as summarizing weekly reports, etc.;

[0235] 3) Classify the content of the target cell, such as classifying different occupations into the education industry, construction industry, etc.;

[0236] 4) Classify the text into different emotions, used for classifying NPS or sentiment of user feedback;

[0237] 5) The user makes a custom question about the cell content, and the large language model answers, which can cover all the above scenarios and has a higher degree of freedom

[0238] Each AI function corresponds to a set of prompts on the server side, and the calculation process of the function is similar to the flowchart shown in the previous part: only the processing process of the front-end configuration interface is bypassed, and the user configures it by himself.

[0239] Taking sentiment analysis as an example, the prompt is as follows:

[0240] "I will provide you with a table of data and a set of labels. Please select the most matching <label> based on the data in the table

[0241] 1. Only select labels from the set of labels, and do not give labels that are not in the set of labels.

[0242] 2. If no matching label is found, return "WPSAI is temporarily unable to generate content. Please try again later"

[0243] 3. Get the label and output it in the following json format:

[0244] {

[0245] "output":<label>

[0246] }

[0247] Example 1:

[0248] Tabular data: This floral dress is really very nice. The small floral patterns are embellished just right, making it very elegant when worn.

[0249] Label set: "Positive, Negative"

[0250] {

[0251] "Output": "Positive"

[0252] }

[0253] The following is the tabular data and label set I give you:

[0254] Tabular data:

[0255] Label set:

[0256] Output: ”.

[0257] The present invention combines AI and formulas elegantly, achieving the intelligent upgrade of formulas. The intelligent application of formulas can significantly reduce manual input; quickly set the AI function columns, reducing repetitive labor; the formulas support flexible parameter customization to meet the needs of different working scenarios. That is, by combining AI and formulas, using the expandable and flexible characteristics of formulas, quickly access the AI capabilities; package the prompt words in different scenarios as different AI capabilities and provide them to users for use, design and expose key parameters for users to adjust; provide users with a front-end interface for batch processing data based on AI formulas, reducing the learning cost and friction for users to configure AI formulas.

[0258] The present invention can be widely applied to the AI content generation and processing functions of tabular data in multiple industries or scenarios such as e-commerce, human resources and administration, customer relationship management, and user feedback.

[0259] We plan to quickly access the large language model into the table in the form of formulas, supplemented by a front-end interaction interface.

[0260] In this embodiment, the text processing ability of the large language model is combined with formulas, and finally a front-end interface for batch processing is delivered to users, improving the efficiency of users in processing text data in the table.

[0261] Next, the data table processing device provided by the present invention will be described. The data table processing device described below can be mutually corresponding and referred to the data table processing method described above.

[0262] Figure 14 is one of the structural schematic diagrams of the data table processing device provided by the present invention, as Figure 14As shown, when applied to a terminal, the data table processing device 1400 includes: a display module 1401, a first determination module 1402, a second determination module 1403, and an addition module 1404, where:

[0263] The display module 1401 is configured to display a configuration page in response to a data configuration instruction for a target sequence in an initial data table;

[0264] The first determination module 1402 is configured to receive configuration parameters based on the configuration page and determine a data processing strategy based on the configuration parameters;

[0265] The second determination module 1403 is configured to determine data to be filled corresponding to the target sequence according to the data processing strategy;

[0266] The addition module 1404 is configured to add the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0267] The data table processing device provided by the present invention displays a configuration page in response to a data configuration instruction for a target sequence in an initial data table; receives configuration parameters based on the configuration page and determines a data processing strategy based on the configuration parameters; determines data to be filled corresponding to the target sequence according to the data processing strategy; and adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table. The present invention generates a data processing strategy through configuration parameters, and then obtains data to be filled, which can accurately and efficiently process user requirements, thereby improving user satisfaction and processing efficiency.

[0268] In one or more alternative embodiments of the present invention, the configuration parameters include a data source area, a processing type, and a processing target;

[0269] The first determination module 1402 is further configured:

[0270] Determine a data filling area according to the target sequence;

[0271] Determine a data processing strategy based on the processing type, the processing target, the data filling area, and the data source area.

[0272] In one or more alternative embodiments of the present invention, the configuration parameters include a data processing strategy;

[0273] The first determination module 1402 is further configured:

[0274] Extract the data processing strategy from the configuration parameters, where the data processing strategy includes a data source area, a processing type, and a processing target.

[0275] In one or more alternative embodiments of the present invention, the second determination module 1403 is further configured to:

[0276] Obtain the data to be processed from the data source area based on the data processing strategy;

[0277] Generate a data processing request according to the data to be processed, the processing type, and the processing target;

[0278] Send the data processing request to the server;

[0279] Receive the data to be filled back by the server based on the data processing request, where the data to be filled is obtained by the server processing the data to be processed based on the processing type and the processing target.

[0280] In one or more alternative embodiments of the present invention, the data processing strategy is a calculation function or a macro;

[0281] The second determination module 1403 is further configured to include:

[0282] In the case where the data processing strategy is a calculation function, determine the data to be filled corresponding to the target sequence according to the calculation function;

[0283] In the case where the data processing strategy is a macro, determine the data to be filled corresponding to the target sequence according to the macro.

[0284] In one or more alternative embodiments of the present invention, the data processing strategy includes at least one of an information extraction strategy, a content summary strategy, a content classification strategy, an emotion recognition strategy, a content translation strategy, and a custom processing strategy;

[0285] The information extraction strategy is used to extract key information from the data source area;

[0286] The content summary strategy is used to summarize the content in the data source area;

[0287] The content classification strategy is used to classify the content in the data source area;

[0288] The emotion recognition strategy is used to recognize the emotion expressed by the content in the data source area;

[0289] The content translation strategy is used to translate the content in the data source area;

[0290] The custom processing strategy is used to perform custom processing on the content in the data source area, and the custom processing is determined based on user requirements.

[0291] In one or more alternative embodiments of the present invention, the information extraction strategy includes a specified information extraction strategy, a multimodal information extraction strategy, and a cross-data source information extraction strategy;

[0292] The specified information extraction strategy is used to extract specified information in the data source area, and the specified information is preset or specified by the user;

[0293] The multimodal information extraction strategy is used to extract information from multimodal content in the data source area, and the multimodal content includes at least one of pictures, attachments, and links;

[0294] The cross-data source information extraction strategy is used to extract specific content from other data sources, and the semantic similarity between the specific content and the selected content is greater than the similarity threshold. The other data sources include other data tables other than the initial data table or other files other than the specified file, and the specified file is the file corresponding to the initial data table.

[0295] In one or more alternative embodiments of the present invention, the first determination module 1402 is further configured to:

[0296] Receive at least two configuration parameters based on the configuration page;

[0297] Determine the sub-data processing strategies corresponding to the respective configuration parameters;

[0298] Fuse the respective sub-data processing strategies to obtain a data processing strategy.

[0299] In one or more alternative embodiments of the present invention, the data table processing device 1400 further includes a second receiving module, which is configured to:

[0300] Receive a selection instruction for the data to be processed in the initial data table, and display a data processing page, where the data processing page includes at least one data processing intention, and the data processing intention is used to jump to the configuration page corresponding to the data processing intention.

[0301] Figure 15 It is the second structural schematic diagram of the data table processing device provided by the present invention. As Figure 15 shown, the data table processing device 1500 includes: a first receiving module 1501, a processing module 1502, and a sending module 1503, where:

[0302] The first receiving module 1501 is configured to receive a data processing request sent by a terminal. The data processing request is generated based on data to be processed, a processing type, and a processing target. The data to be processed, the processing type, and the processing target are determined based on a data processing policy. The data processing policy is obtained from a configuration page for display based on a data configuration instruction for a target sequence in an initial data table.

[0303] The processing module 1502 is configured to process the data to be processed according to the processing type and the processing target to obtain data to be filled.

[0304] The sending module 1503 is configured to send the data to be filled to the terminal, so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0305] The data table processing device provided by the present invention receives a data processing request sent by a terminal. The data processing request is generated based on data to be processed, a processing type, and a processing target. The data to be processed, the processing type, and the processing target are determined based on a data processing policy. The data processing policy is obtained from a configuration page for display based on a data configuration instruction for a target sequence in an initial data table. According to the processing type and the processing target, the data to be processed is processed to obtain data to be filled. The data to be filled is sent to the terminal, so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table. By generating a data processing policy through configuration parameters and then obtaining data to be filled, the present invention can accurately and efficiently process user requirements, thereby improving user satisfaction and processing efficiency.

[0306] In one or more alternative embodiments of the present invention, the processing module 1502 is further configured to:

[0307] Determine model input parameters according to the processing type, the processing target, and the data to be processed;

[0308] Input the model input parameters into an artificial intelligence model for processing to obtain data to be filled.

[0309] In one or more alternative embodiments of the present invention, the model input parameters include a prompt statement, an example, a guiding statement, and sampling parameters. The sampling parameters include an output token parameter and a temperature parameter.

[0310] In one or more alternative embodiments of the present invention, the artificial intelligence model includes a conversion processing layer, a probability prediction layer, and a sampling layer;

[0311] The processing module 1502 is further configured to:

[0312] Convert the prompt statement, the example, and the guiding statement into processing units through the conversion processing layer, and perform data processing on the processing units to obtain data to be output;

[0313] Predict the probability of the data to be output based on the temperature parameter through the probability prediction layer;

[0314] Sample the data to be output based on the output marker parameter and the probability through the sampling layer to obtain data to be filled.

[0315] Figure 16 An example of a schematic physical structure diagram of an electronic device is shown in Figure 16 As shown, the electronic device may include: a processor 1610, a communications interface 1620, a memory 1630, and a communication bus 1640. Among them, the processor 1610, the communications interface 1620, and the memory 1630 communicate with each other through the communication bus 1640. The processor 1610 can call the logical instructions in the memory 1630 to execute a data table processing method, which includes: in response to a data configuration instruction for a target sequence in an initial data table, display a configuration page; receive configuration parameters based on the configuration page, and determine a data processing strategy based on the configuration parameters; determine the data to be filled corresponding to the target sequence according to the data processing strategy; add the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table;

[0316] Alternatively, receive a data processing request sent by a terminal, where the data processing request is generated based on data to be processed, a processing type, and a processing target, the data to be processed, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from a configuration page displayed in response to a data configuration instruction for a target sequence in an initial data table; process the data to be processed according to the processing type and the processing target to obtain data to be filled; send the data to be filled to the terminal so that the terminal adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0317] In addition, when the logical instructions in the aforementioned memory 1630 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0318] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the data table processing method provided by the above-mentioned various methods. The method includes: in response to a data configuration instruction for a target sequence in an initial data table, presenting a configuration page; receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters; determining the data to be filled corresponding to the target sequence according to the data processing strategy; adding the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0319] Alternatively, receiving a data processing request sent by a terminal, the data processing request is generated based on data to be processed, a processing type, and a processing target. The data to be processed, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from the configuration page presented in response to a data configuration instruction for a target sequence in an initial data table; processing the data to be processed according to the processing type and the processing target to obtain data to be filled; sending the data to be filled to the terminal so that the terminal adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0320] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a data table processing method provided by the above-mentioned various methods. The method includes: in response to a data configuration instruction for a target sequence in an initial data table, displaying a configuration page; receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters; determining to-be-filled data corresponding to the target sequence according to the data processing strategy; adding the to-be-filled data to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table;

[0321] Alternatively, receiving a data processing request sent by a terminal, the data processing request is generated based on to-be-processed data, a processing type, and a processing target, the to-be-processed data, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from a configuration page displayed in response to a data configuration instruction for a target sequence in an initial data table; processing the to-be-processed data according to the processing type and the processing target to obtain to-be-filled data; sending the to-be-filled data to the terminal so that the terminal adds the to-be-filled data to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

[0322] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0323] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0324] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for processing data tables, characterized in that, applied to a terminal, including: Responding to a data configuration instruction for a target sequence in an initial data table, and presenting a configuration page; Receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters; Determining the data to be filled corresponding to the target sequence according to the data processing strategy; Adding the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

2. The data table processing method according to claim 1, characterized in that, The configuration parameters include a data source area, a processing type, and a processing target; The determining a data processing strategy based on the configuration parameters includes: Determining a data filling area according to the target sequence; Determining a data processing strategy based on the processing type, the processing target, the data filling area, and the data source area.

3. The data table processing method according to claim 1, characterized in that, The configuration parameters include a data processing strategy; The determining a data processing strategy based on the configuration parameters includes: Extracting the data processing strategy from the configuration parameters, and the data processing strategy includes a data source area, a processing type, and a processing target.

4. The data table processing method according to claim 2 or 3, characterized in that, The determining the data to be filled corresponding to the target sequence according to the data processing strategy includes: Obtaining data to be processed from the data source area based on the data processing strategy; Generating a data processing request according to the data to be processed, the processing type, and the processing target; Sending the data processing request to the server; Receiving the data to be filled fed back by the server based on the data processing request, and the data to be filled is obtained by the server processing the data to be processed based on the processing type and the processing target.

5. The data table processing method according to claim 1, characterized in that, The data processing strategy is a calculation function or a macro; The determining the data to be filled corresponding to the target sequence according to the data processing strategy includes: In the case where the data processing strategy is a calculation function, determining the data to be filled corresponding to the target sequence according to the calculation function; In the case where the data processing strategy is a macro, determining the data to be filled corresponding to the target sequence according to the macro.

6. The data table processing method according to claim 1, characterized in that, The data processing strategy includes at least one of an information extraction strategy, a content summary strategy, a content classification strategy, an emotion recognition strategy, a content translation strategy, and a custom processing strategy; The information extraction strategy is used to extract key information in the data source area; The content summary strategy is used to summarize the content in the data source area; The content classification strategy is used to classify the content in the data source area; The emotion recognition strategy is used to recognize the emotion expressed by the content in the data source area; The content translation strategy is used to translate the content in the data source area; The custom processing strategy is used to perform custom processing on the content in the data source area, and the custom processing is determined based on user requirements.

7. The data table processing method according to claim 6, wherein, the information extraction strategy includes a specified information extraction strategy, a multimodal information extraction strategy, and a cross-data-source information extraction strategy; the specified information extraction strategy is used to extract specified information in the data source area, and the specified information is preset or specified by the user; the multimodal information extraction strategy is used to extract information from multimodal content in the data source area, and the multimodal content includes at least one of pictures, attachments, and links; the cross-data-source information extraction strategy is used to extract specific content from other data sources, and the semantic similarity between the specific content and the selected content is greater than the similarity threshold. The other data sources include other data tables other than the initial data table or other files other than the specified file, and the specified file is the file corresponding to the initial data table.

8. The data table processing method according to any one of claims 1-3, wherein, receiving configuration parameters based on the configuration page, and determining a data processing strategy based on the configuration parameters, includes: receiving at least two configuration parameters based on the configuration page; determining sub-data processing strategies respectively corresponding to the configuration parameters; fusing the sub-data processing strategies to obtain a data processing strategy.

9. The data table processing method according to any one of claims 1-3, wherein, the method further includes: receiving a selection instruction for the data to be processed in the initial data table, and displaying a data processing page, where the data processing page includes at least one data processing intention, and the data processing intention is used to jump to the configuration page corresponding to the data processing intention.

10. A data table processing method, wherein, applied to a server, includes: receiving a data processing request sent by a terminal, where the data processing request is generated based on data to be processed, a processing type, and a processing target, and the data to be processed, the processing type, and the processing target are determined based on a data processing strategy, and the data processing strategy is obtained from a configuration page displayed based on a data configuration instruction for a target sequence in an initial data table; processing the data to be processed according to the processing type and the processing target to obtain data to be filled; sending the data to be filled to the terminal, so that the terminal adds the data to be filled to the data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

11. The data table processing method according to claim 10, wherein, processing the data to be processed according to the processing type and the processing target to obtain data to be filled includes: determining model input parameters according to the processing type, the processing target, and the data to be processed; inputting the model input parameters into an artificial intelligence model for processing to obtain data to be filled.

12. The data table processing method according to claim 11, wherein, The model input parameters include a prompt statement, examples, a guiding statement, and sampling parameters, and the sampling parameters include an output token parameter and a temperature parameter.

13. The data table processing method according to claim 12, wherein, the artificial intelligence model includes a conversion processing layer, a probability prediction layer, and a sampling layer; the step of inputting the model input parameters into the artificial intelligence model for processing to obtain data to be filled includes: through the conversion processing layer, converting the prompt statement, the examples, and the guiding statement into processing units, and performing data processing on the processing units to obtain data to be output; through the probability prediction layer, predicting the probability of the data to be output based on the temperature parameter; through the sampling layer, sampling the data to be output based on the output token parameter and the probability to obtain data to be filled.

14. A data table processing device, wherein, applied to a terminal, and includes: a display module configured to display a configuration page in response to a data configuration instruction for a target sequence in an initial data table; a first determination module configured to receive configuration parameters based on the configuration page and determine a data processing strategy based on the configuration parameters; a second determination module configured to determine data to be filled corresponding to the target sequence according to the data processing strategy; an adding module configured to add the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

15. A data table processing device, wherein, applied to a server, and includes: a first receiving module configured to receive a data processing request sent by a terminal, the data processing request being generated based on data to be processed, a processing type, and a processing target, the data to be processed, the processing type, and the processing target being determined based on a data processing strategy, and the data processing strategy being obtained from a configuration page displayed in response to a data configuration instruction for a target sequence in an initial data table; a processing module configured to process the data to be processed according to the processing type and the processing target to obtain data to be filled; a sending module configured to send the data to be filled to the terminal so that the terminal adds the data to be filled to a data filling area corresponding to the target sequence in the initial data table to obtain a target data table.

16. An electronic device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, when the processor executes the program, it implements the data table processing method according to any one of claims 1 to 9 or 10 to 13.

17. A computer-readable storage medium, on which a computer program is stored, wherein, when the computer program is executed by a processor, it implements the data table processing method according to any one of claims 1 to 9 or 10 to 13.

Images