Form data visualization generation method and system based on Data Science

Abstract

The invention discloses a form data visualization generation method and system based on Data Science, belongs to the technical field of data visualization, and aims to solve the technical problem of how to provide whole-system-oriented data visualization and data analysis for a user, improve the data visualization experience of the user and improve the user experience. According to the technical scheme, the method comprises the steps that data preprocessing is carried out, wherein data preprocessing is carried out on fields and data of a form separately, and preprocessed data is obtained; feature extraction: utilizing a Glove model to perform word embedding feature extraction on the preprocessed data by constructing a vocabulary; predicting a data topic: predicting a topic result through an SVM model; visual parameters are generated, wherein parameters used for data visualization are generated through a Transform model; and data visualization is carried out, wherein data visualization is realized through a numpy library, a matplotlib library or a search library.

Description

Technical Field

[0001] The present invention relates to the technical field of data visualization, and in particular to a form data visualization generation method and system based on Data Science. Background Art

[0002] Data visualization, as a visual art of data interpretation, integrates multidisciplinary technologies such as graphic design, image processing, computer vision, and user interface design. It effectively conveys information and reveals the deep meaning and complex patterns behind the data by converting data into intuitive and easy-to-understand graphics and images. Data visualization not only provides professionals with a unique perspective to gain insight into data sets, but also brings an intuitive and clear data cognition experience to ordinary users.

[0003] In the field of data analysis, data visualization plays a vital role in both business analysis (BA) closely integrated with the business and scientific research based on machine learning (ML). Compared with boring text reports, data visualization charts can more intuitively show the full picture of the data and help users quickly capture key information. Therefore, regardless of the form, data visualization modules have become an indispensable part of platform and system design. These modules are usually used to display statistical data such as "total data", "amount of subject data" and "number of system users". The data obtained from the back end is simply processed and presented through carefully selected visualization elements on the front end. However, this traditional data visualization method is often limited by the scope of display, has high development costs, and requires continuous functional maintenance in the later stage.

[0004] For complex systems, especially those containing multiple functional forms and ledgers, if users want to visualize the data in each form, the above development model is particularly cumbersome, and the visualization form is too single. The current solution is to use professional third-party data visualization tools such as Tableau and ChartBlocks to generate various forms of visualization, such as static, dynamic and interactive. However, using such tools requires manual table data download, tool use, and business judgment to select the appropriate visualization form, which is not easy to operate and implement.

[0005] Therefore, how to provide users with data visualization and data analysis for the entire system and improve users' data visualization experience is a technical problem that needs to be solved urgently. Summary of the invention

[0006] The technical task of the present invention is to provide a form data visualization generation method and system based on Data Science to solve the problem of how to provide users with system-wide data visualization and data analysis and enhance users' data visualization experience.

[0007] The technical task of the present invention is achieved in the following way: a form data visualization generation method based on Data Science, the method is as follows:

[0008] Data preprocessing: preprocess the fields and data of the form separately to obtain the preprocessed data;

[0009] Feature extraction: Use the Glove model to extract word embedding features from the preprocessed data by building a vocabulary;

[0010] Predict data topics: predict topic results through the SVM (Support Vector Machine) model;

[0011] Generate visualization parameters: Generate parameters for data visualization through the Transformer model;

[0012] Data visualization: Data visualization is achieved through numpy, matplotlib or searbone libraries.

[0013] Preferably, the fields and data of the form are separately preprocessed as follows:

[0014] Processing of form fields: converting categorical data into numerical data using Label-Encoding, and then scaling all numerical data features through standardization and normalization to make all numerical data features within a similar range; using the processed features to create new data features for existing features by combining features and polynomial features, and thus extracting more useful information;

[0015] Processing of form data: processing missing values, outliers, and non-standard data that does not conform to the field type, and also performing numerical conversion; if the data volume of the data set is larger than the preset number, use the Data Normalization method of clustering, dimensionality reduction and sampling to reduce the data volume while maintaining the integrity of the original data.

[0016] Preferably, the GloVe model is used to extract word embedding features from the preprocessed data by constructing a vocabulary table as follows:

[0017] Take the preprocessed data as the input set, traverse the input set, count the frequency of each word co-occurring with the corresponding context word, and build a huge context co-occurrence matrix;

[0018] Use the GloVe model to factorize the word-context co-occurrence matrix and learn the word embedding representation by minimizing the reconstruction error;

[0019] Generate a vocabulary representing the feature values ​​of the form data and the corresponding word vector matrix.

[0020] Preferably, the topic prediction results using the SVM model are as follows:

[0021] The input parameter of the SVM (support vector machine) model is the feature vector. The training data set of the SVM model uses the form data of different industries (if there is a targeted industry development demand, the data proportion of the industry will be increased). The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters (C value, kernel function type, etc.), and the test set is used to evaluate the prediction accuracy of the final model.

[0022] After the SVM model is trained, the data set of the form data feature values ​​is converted into a feature vector and input into the SVM model to obtain the predicted topic keyword results.

[0023] Preferably, the visualization parameters are generated as follows:

[0024] The prediction topic keywords of the SVM model and the field names that users can select (such as the quantity of goods and yesterday's business flow) are used as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model has a good generalization ability for processing sequence data by capturing the bidirectional dependency information of the input data;

[0025] The input values ​​are processed to generate a paragraph describing how to generate visualizations, for example: "The visualization of this form should be drawn using the boxplox function of the seaborn library to highlight the median, quartiles, and outliers; x is the category, y is the price, bins = 30, kde = True", corresponding to the parameters for generating the visualization.

[0026] Preferably, the data visualization is as follows:

[0027] According to the visualization parameter information, the API is automatically called or the visualization library is drawn. After the visualization parameter information is input, the visualization solution of the corresponding form is obtained, and the data analysis solution for the corresponding visualization data is obtained through the GPT model.

[0028] A form data visualization generation system based on Data Science, the system comprising:

[0029] The data preprocessing module is used to preprocess the fields and data of the form separately and obtain the preprocessed data;

[0030] The feature extraction module is used to extract word embedding features from the preprocessed data by building a vocabulary using the Glove model;

[0031] Data topic prediction module, used to predict topic results through SVM (Support Vector Machine) model;

[0032] Visualization parameter generation module, used to generate parameters for data visualization through the Transformer model;

[0033] Data visualization module for data visualization through numpy, matplotlib or searbone library.

[0034] Preferably, the data preprocessing module comprises:

[0035] The submodule for processing the form fields is used to convert categorical data into numerical data using Label-Encoding, and then scale all numerical data features through standardization and normalization to make all numerical data features within a similar range; the processed features are combined with existing features to create new data features, thereby extracting more useful information;

[0036] The form data processing submodule is used to process missing values, abnormal values, and non-standard data that does not conform to the field type, and also performs numerical conversion. If the data volume of the data set is larger than the preset number, the Data Normalization method of clustering, dimensionality reduction and sampling is used to reduce the data volume while maintaining the integrity of the original data.

[0037] The feature extraction module comprises:

[0038] The context co-occurrence matrix construction submodule is used to take the pre-processed data as the input set, traverse the input set, count the frequency of each word co-occurring with the corresponding context word, and construct a huge context co-occurrence matrix;

[0039] The embedding reconstruction submodule is used to factorize the word-context co-occurrence matrix using the GloVe model and learn the embedding representation of words by minimizing the reconstruction error;

[0040] The vocabulary and word embedding matrix generation submodule is used to generate a vocabulary representing the feature values ​​of the form data and the corresponding word embedding matrix;

[0041] The data topic prediction module includes:

[0042] The model training submodule is used to set the input parameters of the SVM (support vector machine) model as feature vectors. The training data set of the SVM model uses the form data of different industries (if there is a targeted industry development demand, the data proportion of the industry will be increased). The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters (C value, kernel function type, etc.), and the test set is used to evaluate the prediction accuracy of the final model.

[0043] The topic keyword acquisition submodule is used to convert the data set of form data feature values ​​into feature vectors after the SVM model is trained, and input them into the SVM model to obtain the predicted topic keyword results;

[0044] The visualization parameter generation module includes:

[0045] The capture submodule is used to use the prediction topic keywords of the SVM model and the field names that can be selected by the user (such as the quantity of goods and yesterday's business flow) as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model has a good generalization ability for processing sequence data by capturing the bidirectional dependency information of the input data;

[0046] The visualization text and parameter generation submodule is used to process the input value to generate a paragraph of text that describes how to generate visualization, for example: "The visualization of this form should be drawn using the boxplox function of the seaborn library to highlight the median, quartiles, and outliers; x is the category, y is the price, bins = 30, kde = True", and the corresponding parameters for generating visualization;

[0047] The data visualization module automatically calls the API or draws the visualization library according to the visualization parameter information, obtains the visualization scheme of the corresponding form after inputting the visualization parameter information, and obtains the data analysis scheme for the corresponding visualization data through the GPT model.

[0048] An electronic device comprising: a memory and at least one processor;

[0049] Wherein, the memory stores a computer program;

[0050] The at least one processor executes the computer program stored in the memory, so that the at least one processor performs the above-mentioned form data visualization generation method based on Data Science.

[0051] A computer-readable storage medium stores a computer program, which can be executed by a processor to implement the above-mentioned form data visualization generation method based on Data Science.

[0052] The Data Science-based form data visualization generation method and system of the present invention have the following advantages:

[0053] (1) The data visualization method technology of the form generation based on Data Science technology of the present invention is based on the relevant knowledge theory and model methods of natural language processing, machine learning, and deep learning. Through data preprocessing, data feature extraction, and data theme prediction, the method library and parameters for executing data visualization operations are obtained. It can be widely used in various industries and provide strong technical support for data analysis related functions;

[0054] (ii) The Glove (Global Vectors for Word Representation) algorithm of the present invention is used to extract data features, the SVM (Support Vector Machine) model predicts topic results, and the Transformer+GPT model generates parameters for data visualization. Finally, data visualization is achieved through libraries such as numpy, matplotlib, and searbon;

[0055] (III) The present invention uses the idea of ​​automatic data visualization provided by large models such as GPT, and uses technologies and models related to the Data Science discipline to extract data features from fields and data in the form, automatically generates parameters for building data visualization to generate visualization and provide data analysis text. It can not only provide users with data visualization and data analysis for the entire system, but also generate a visual feast of different dimensions and modes for a single form based on the user's personal data preferences;

[0056] (IV) The present invention is an enhancement of the data analysis function in various industry systems. On a macro level, it provides the system with more professional and institutionalized BA data analysis and visualization in various styles that fit the business of the industry. On a micro level, it is object-oriented and can be as small as a data table in each system. When the user needs it, it provides data visualization and data analysis description for the table. It uses the language described by the GPT model to avoid excessive professionalism and is easy to understand without losing content.

[0057] (V) The present invention creates different values ​​in various fields. For internal management of enterprises, the present invention helps to achieve refined management of internal data. Through automated data visualization and analysis, it eliminates the tedious daily business flow reports, personnel attendance statistics and other internal management tasks. For companies with higher financial and economic value, selecting parameters that are more inclined to the financial field can provide them with an efficient BA analysis tool, which can substantially reduce both labor costs and time costs, bringing value to various industries, fields and institutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The present invention is further described below in conjunction with the accompanying drawings.

[0059] Attached Figure 1 Schematic diagram of the form data visualization generation method based on Data Science. DETAILED DESCRIPTION

[0060] The Data Science-based form data visualization generation method and system of the present invention are described in detail below with reference to the accompanying drawings and specific embodiments of the specification.

[0061] Embodiment 1:

[0062] As attached Figure 1 As shown, the form data visualization generation method based on Data Science of the present invention is as follows:

[0063] S1. Data preprocessing: preprocess the fields and data of the form separately to obtain the preprocessed data;

[0064] S2, Feature Extraction: Use the Glove model to extract word embedding features from the preprocessed data by building a vocabulary;

[0065] S3. Predict data topics: predict topic results through the SVM (Support Vector Machine) model;

[0066] S4, Generate visualization parameters: Generate parameters for data visualization through the Transformer model;

[0067] S5, data visualization: data visualization is realized by using numpy, matplotlib or searbone library. In this embodiment, the data preprocessing of the fields and data of the form in step S1 is as follows:

[0068] S101. Processing the fields of the form: converting categorical data into numerical data using Label-Encoding, and then scaling all numerical data features through standardization and normalization to make all numerical data features within a similar range; using the processed features to create new data features for existing features by combining features and polynomial features, and then extracting more useful information;

[0069] S102. Processing of form data: processing missing values, abnormal values, and processing of non-standard data that does not conform to the field type, and performing numerical conversion on the same; if the data volume of the data set is larger than the preset number, use the Data Normalization method of clustering, dimensionality reduction and sampling to reduce the data volume while nearly maintaining the integrity of the original data.

[0070] In step S2 of this embodiment, the word embedding feature extraction of the preprocessed data by using the GloVe model by constructing a vocabulary is specifically as follows:

[0071] S201, taking the preprocessed data as an input set, traversing the input set, counting the frequency of co-occurrence of each word with the corresponding context word, and constructing a huge context co-occurrence matrix;

[0072] S202, factorizing the word-context co-occurrence matrix using the GloVe model, and learning the embedding representation of the word by minimizing the reconstruction error;

[0073] S203, generating a vocabulary representing feature values ​​of the form data and a corresponding word vector matrix (embedding matrix).

[0074] The topic prediction results by the SVM model in step S3 of this embodiment are as follows:

[0075] S301. The input parameter of the SVM (support vector machine) model is a feature vector. The training data set of the SVM model uses the form data of different industries (if there is a targeted industry development demand, the data proportion of the industry will be increased). The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters (C value, kernel function type, etc.), and the test set is used to evaluate the prediction accuracy of the final model.

[0076] S302, after the SVM model is trained, the data set of the form data feature values ​​is converted into a feature vector, which is input into the SVM model to obtain the predicted topic keyword results.

[0077] The specific visualization parameters generated in step S4 of this embodiment are as follows:

[0078] S401, the prediction topic keywords of the SVM model, and the field names that can be selected by the user (such as the quantity of goods, yesterday's business flow) are used as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model has a good generalization ability for processing sequence data by capturing the bidirectional dependency information of the input data;

[0079] S402, processing the input value to generate a paragraph of text for describing how to generate visualization, for example: "The visualization of this form should be drawn using the boxplox function of the seaborn library to highlight the median, quartiles, and outliers; x is the category, y is the price, bins = 30, kde = True", and corresponding parameters for generating the visualization.

[0080] The data visualization in step S5 of this embodiment is specifically as follows:

[0081] According to the visualization parameter information, the API is automatically called or the visualization library is drawn. After the visualization parameter information is input, the visualization solution of the corresponding form is obtained, and the data analysis solution for the corresponding visualization data is obtained through the GPT model.

[0082] Embodiment 2:

[0083] This embodiment provides a form data visualization generation system based on Data Science, the system comprising:

[0084] The data preprocessing module is used to preprocess the fields and data of the form separately and obtain the preprocessed data;

[0085] The feature extraction module is used to extract word embedding features from the preprocessed data by building a vocabulary using the Glove model;

[0086] Data topic prediction module, used to predict topic results through SVM (Support Vector Machine) model;

[0087] Visualization parameter generation module, used to generate parameters for data visualization through the Transformer model;

[0088] Data visualization module for data visualization through numpy, matplotlib or searbone library.

[0089] The data preprocessing module in this embodiment includes:

[0090] The submodule for processing the form fields is used to convert categorical data into numerical data using Label-Encoding, and then scale all numerical data features through standardization and normalization to make all numerical data features within a similar range; the processed features are combined with existing features to create new data features, thereby extracting more useful information;

[0091] The form data processing submodule is used to handle missing values, abnormal values, and non-standard data that does not conform to the field type, and also performs numerical conversion; if the data volume of the data set is larger than the preset number, the Data Normalization method of clustering, dimensionality reduction and sampling is used to reduce the data volume while nearly maintaining the integrity of the original data.

[0092] The feature extraction module in this embodiment includes:

[0093] The context co-occurrence matrix construction submodule is used to take the pre-processed data as the input set, traverse the input set, count the frequency of each word co-occurring with the corresponding context word, and construct a huge context co-occurrence matrix;

[0094] The embedding reconstruction submodule is used to factorize the word-context co-occurrence matrix using the GloVe model and learn the embedding representation of words by minimizing the reconstruction error;

[0095] The vocabulary and word embedding matrix generation submodule is used to generate a vocabulary (vocabulary) representing the feature values ​​of the form data and the corresponding word embedding matrix (embedding matrix).

[0096] The data topic prediction module in this embodiment includes:

[0097] The model training submodule is used to set the input parameters of the SVM (support vector machine) model as feature vectors. The training data set of the SVM model uses the form data of different industries (if there is a targeted industry development demand, the data proportion of the industry will be increased). The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters (C value, kernel function type, etc.), and the test set is used to evaluate the prediction accuracy of the final model.

[0098] The topic keyword acquisition submodule is used to convert the data set of form data feature values ​​into feature vectors after the SVM model is trained, and input them into the SVM model to obtain the predicted topic keyword results.

[0099] The visualization parameter generation module in this embodiment includes:

[0100] The capture submodule is used to use the prediction topic keywords of the SVM model and the field names that can be selected by the user (such as the quantity of goods and yesterday's business flow) as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model has a good generalization ability for processing sequence data by capturing the bidirectional dependency information of the input data;

[0101] The visualization text and parameter generation submodule is used to process the input value to generate a paragraph of text that describes how to generate the visualization, for example: "The visualization of this form should be drawn using the boxplox function of the seaborn library to highlight the median, quartiles, and outliers; x is the category, y is the price, bins = 30, kde = True", and the corresponding parameters for generating the visualization.

[0102] The data visualization module in this embodiment automatically calls the API or draws the visualization library according to the visualization parameter information, obtains the visualization solution of the corresponding form after inputting the visualization parameter information, and obtains the data analysis solution for the corresponding visualization data through the GPT model.

[0103] Embodiment 3:

[0104] This embodiment also provides an electronic device, including: a memory and a processor;

[0105] Wherein, the memory stores computer-executable instructions;

[0106] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the form data visualization generation method based on Data Science in any embodiment of the present invention.

[0107] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor may be a microprocessor or any conventional processor, etc.

[0108] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the terminal, etc. In addition, the memory can also include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash memory card, at least one disk storage period, a flash memory device, or other volatile solid-state storage devices.

[0109] Embodiment 4:

[0110] This embodiment also provides a computer-readable storage medium, which stores a plurality of instructions, which are loaded by a processor, so that the processor executes the form data visualization generation method based on Data Science in any embodiment of the present invention. Specifically, a system or device equipped with a storage medium can be provided, on which a software program code that implements the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0111] In this case, the program code itself read from the storage medium can realize the function of any one of the above-mentioned embodiments, and thus the program code and the storage medium storing the program code constitute a part of the present invention.

[0112] The storage medium embodiments for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RYM, DVD-RW, DVD+RW), a magnetic tape, a non-volatile memory card, and a ROM. Alternatively, the program code can be downloaded from a server computer via a communication network.

[0113] In addition, it should be clear that the functions of any of the above embodiments can be implemented not only by executing the program code read by the computer, but also by enabling an operating system operating on the computer to complete part or all of the actual operations based on instructions from the program code.

[0114] In addition, it can be understood that the program code read from the storage medium is written to a memory provided in an expansion board inserted into the computer or written to a memory provided in an expansion unit connected to the computer, and then based on the instructions of the program code, a CPU installed on the expansion board or the expansion unit is enabled to perform part or all of the actual operations, thereby realizing the functions of any of the above-mentioned embodiments.

[0115] 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 aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A form data visualization generation method based on Data Science, characterized in that: The method is as follows: Data preprocessing: preprocess the fields and data of the form separately to obtain the preprocessed data; Feature extraction: Use the Glove model to extract word embedding features from the preprocessed data by building a vocabulary; Predict data topics: predict topic results through SVM model; Generate visualization parameters: Generate parameters for data visualization through the Transformer model; Data visualization: Data visualization is achieved through numpy, matplotlib or searbone libraries.

2. The form data visualization generation method based on Data Science according to claim 1 is characterized in that: The data preprocessing for the form fields and data is as follows: Processing of form fields: converting categorical data into numerical data using Label-Encoding, and then scaling all numerical data features through standardization and normalization to make all numerical data features within a similar range; using the processed features to create new data features for existing features by combining features and polynomial features, and thus extracting more useful information; Processing of form data: processing missing values, outliers, and non-standard data that does not conform to the field type, and also performing numerical conversion; if the data volume of the data set is larger than the preset number, use the Data Normalization method of clustering, dimensionality reduction and sampling to reduce the data volume while maintaining the integrity of the original data.

3. The form data visualization generation method based on Data Science according to claim 1 or 2, characterized in that: The GloVe model is used to extract word embedding features from preprocessed data by building a vocabulary table as follows: Take the preprocessed data as the input set, traverse the input set, count the frequency of each word co-occurring with the corresponding context word, and build a context co-occurrence matrix; Use the GloVe model to factorize the word-context co-occurrence matrix and learn the word embedding representation by minimizing the reconstruction error; Generate a vocabulary table representing the feature values ​​of the form data and the corresponding word vector matrix.

4. The form data visualization generation method based on Data Science according to claim 3 is characterized in that: The results of predicting topics through the SVM model are as follows: The input parameter of the SVM model is a feature vector. The training data set of the SVM model uses form data from different industries. The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters, and the test set is used to evaluate the prediction accuracy of the final model. After the SVM model is trained, the data set of the form data feature values ​​is converted into a feature vector and input into the SVM model to obtain the predicted topic keyword results.

5. The form data visualization generation method based on Data Science according to claim 4 is characterized in that: The visualization parameters are as follows: The predicted topic keywords of the SVM model and the field names that can be selected by the user are used as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model captures the bidirectional dependency information of the input data; The input value is processed to generate a text describing how to generate the visualization, and the corresponding parameters for drawing the visualization are generated.

6. The form data visualization generation method based on Data Science according to claim 5 is characterized in that: The data visualization is as follows: According to the visualization parameter information, the API is automatically called or the visualization library is drawn. After the visualization parameter information is input, the visualization solution of the corresponding form is obtained, and the data analysis solution for the corresponding visualization data is obtained through the GPT model.

7. A form data visualization generation system based on Data Science, characterized in that: The system includes: The data preprocessing module is used to preprocess the fields and data of the form separately and obtain the preprocessed data; The feature extraction module is used to extract word embedding features from the preprocessed data by building a vocabulary using the Glove model; Data topic prediction module, used to predict topic results through SVM model; Visualization parameter generation module, used to generate parameters for data visualization through the Transformer model; Data visualization module for data visualization through numpy, matplotlib or searbone library.

8. The form data visualization generation system based on Data Science according to claim 7, characterized in that: The data preprocessing module comprises: The submodule for processing the form fields is used to convert categorical data into numerical data using Label-Encoding, and then scale all numerical data features through standardization and normalization to make all numerical data features within a similar range; the processed features are combined with existing features to create new data features, thereby extracting more useful information; The form data processing submodule is used to process missing values, abnormal values, and non-standard data that does not conform to the field type, and also performs numerical conversion. If the data volume of the data set is larger than the preset number, the Data Normalization method of clustering, dimensionality reduction and sampling is used to reduce the data volume while maintaining the integrity of the original data. The feature extraction module comprises: The context co-occurrence matrix construction submodule is used to take the pre-processed data as the input set, traverse the input set, count the frequency of co-occurrence of each word with the corresponding context word, and construct a context co-occurrence matrix; The embedding representation reconstruction submodule is used to factorize the word-context co-occurrence matrix using the GloVe model and learn the embedding representation of words by minimizing the reconstruction error; The vocabulary and word vector matrix generation submodule is used to generate a vocabulary representing the feature values ​​of the form data and the corresponding word vector matrix; The data topic prediction module includes: The model training submodule is used to set the input parameters of the SVM model as feature vectors. The training data set of the SVM model uses the form data of different industries. The training data set is divided into a training set, a validation set, and a test set. The training set is used to train the SVM model, the validation set is used to adjust the parameters, and the test set is used to evaluate the prediction accuracy of the final model. The topic keyword acquisition submodule is used to convert the data set of form data feature values ​​into feature vectors after the SVM model is trained, and input them into the SVM model to obtain the predicted topic keyword results; The visualization parameter generation module includes: The capture submodule is used to use the predicted topic keywords of the SVM model and the field names that can be selected by the user as the input values ​​of the Transformer model. The bidirectional information transmission mechanism of the Transformer model captures the bidirectional dependency information of the input data; The visualization text and parameter generation submodule is used to process the input value to generate a text describing how to generate visualization, and correspondingly generate the parameters for drawing visualization; The data visualization module automatically calls the API or draws the visualization library according to the visualization parameter information, obtains the visualization scheme of the corresponding form after inputting the visualization parameter information, and obtains the data analysis scheme for the corresponding visualization data through the GPT model.

9. An electronic device, characterized in that: include: memory and at least one processor; Wherein, the memory stores a computer program; The at least one processor executes the computer program stored in the memory, so that the at least one processor performs the form data visualization generation method based on Data Science according to any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which can be executed by a processor to implement the form data visualization generation method based on DataScience as described in any one of claims 1 to 6.

Images