Large-model-based data large-screen visualization method, apparatus and device, and medium

Abstract

The invention discloses a data large screen visualization method and device based on a large model, equipment and a medium, and relates to the technical field of data large screens, and the method comprises the steps: obtaining externally input user voice data, carrying out the intention analysis of the user voice data, and determining a query prompt word representing the intention from the user voice data; inputting the query prompt word into a preset large model for analysis, and generating target data corresponding to the query prompt word and a target chart template matched with the query prompt word; and mapping the target data to the target chart template, determining a target chart, and visualizing the target chart. According to the method, the large-screen data display efficiency can be improved.

Description

Technical Field

[0001] The present application relates to the field of data large screen technology, and in particular to a data large screen visualization method, device, equipment and medium based on a large model. Background Art

[0002] With the advent of the big data era, data visualization has become one of the important means for enterprises to make efficient decisions.

[0003] As a way to intuitively display large amounts of data, data screens have been widely used in real-time monitoring, business analysis and other fields. However, the current data screen technology still has certain limitations. Most data screens can only display preset chart types, such as bar charts, line charts, etc. The selection of these charts is often determined during the development phase, resulting in fixed display content on the data screen, making it difficult to quickly adjust the displayed content and format according to actual needs. Once the data changes or new business needs arise, the code needs to be reconfigured or even rewritten to achieve updates, which not only increases the workload but also reduces flexibility.

[0004] Therefore, how to solve the problem that changes in business needs require the large screen to be reconfigured for visualization is a problem that needs to be solved urgently.

[0005] The above contents are only used to assist in understanding the technical solution of the present application and do not constitute an admission that the above contents are prior art. Summary of the invention

[0006] The main purpose of this application is to provide a large-screen data visualization method, device, equipment and storage medium based on a large model, aiming to solve the technical problem that the large screen needs to be reconfigured for visualization due to changes in business requirements.

[0007] To achieve the above purpose, the present application proposes a large-screen data visualization method based on a large model, and the large-screen data visualization method based on a large model includes: Acquire externally input user voice data, perform intent analysis on the user voice data, and determine a query prompt word representing the intent from the user voice data; Input the query prompt word into a preset large model for analysis, and generate target data corresponding to the query prompt word and a target chart template matching the query prompt word; The target data is mapped to the target chart template, a target chart is determined, and the target chart is visualized.

[0008] In one embodiment, the step of performing intent analysis on the user voice data and determining a query prompt word from the user voice data includes: Converting the user voice data into text data; Performing intent analysis on the text data to determine user intent; Identifying entity keywords in the text data, wherein the entity keywords include one or more of topic keywords, time range keywords, and location range keywords; According to the user intention and the entity keyword, the entity keyword is converted from colloquial language to a standardized query prompt word representing the user intention.

[0009] In one embodiment, the step of inputting the query prompt word into a preset large model for analysis to generate target data corresponding to the query prompt word and a target chart template matching the query prompt word includes: Input the query prompt word into the preset large model, and obtain the database query statement and the chart template generation statement output by the preset large model that match the query prompt word; The target data is acquired according to the database query statement, and a target chart template is generated according to the chart template generation statement.

[0010] In one embodiment, the step of inputting the query prompt word into the preset large model and obtaining the database query statement and the chart template generation statement output by the preset large model includes: Input the query prompt word into the preset large model, and obtain a database query statement output by the preset large model, wherein the preset large model determines a data set matching the query prompt word in a preset database based on the query prompt word, and generates and outputs a database query statement corresponding to the data set; Obtain a preset chart template, input each of the preset chart templates and the query prompt word into the preset big model, and obtain a chart template generation statement output by the preset big model, wherein the preset big model matches each of the preset chart templates based on the query prompt word, determines a successfully matched preset chart template, and generates a chart template generation statement for the successfully matched preset chart template according to the query prompt word.

[0011] In one embodiment, before the step of inputting each of the preset chart templates and the query prompt words into the preset macro model, the method further includes: Create preset chart templates for different chart types; Construct a template description corresponding to each of the preset chart templates, wherein the template description includes applicable data type, creator, creation date, version, function description, classification label and application scenario.

[0012] In one embodiment, the step of inputting each of the preset chart templates and the query prompt words into the preset macro model to obtain the chart template generation statement output by the preset macro model includes: The query prompt word is input into the preset big model, and the chart template generation statement generated by the preset big model is obtained, wherein the preset big model determines the data type of the data set according to the data set matching the query prompt word in the preset database, and matches the template description of each preset chart template according to the data type of the data set and the query prompt word, determines the chart template that matches successfully, and generates the chart template generation statement of the chart template that matches successfully.

[0013] In one embodiment, before the step of inputting the query prompt word into a preset large model for analysis, the method further includes: Generate a chart dataset based on a preset multimodal language model; Fine-tune the preset large model using the chart data set to determine the fine-tuned preset large model; The step of inputting each of the preset chart templates and the query prompt words into the preset large model to obtain the chart template generation statement output by the preset large model includes: Input each of the preset chart templates and the query prompt word into a fine-tuned preset large model, and obtain a chart template generation statement generated by the fine-tuned preset large model, wherein the fine-tuned preset large model determines the data type of the data set according to the data set matching the query prompt word in a preset database, and matches the template description of each of the preset chart templates according to the data type of the data set and the query prompt word to determine a target chart template that has been successfully matched, determines the template parameters of the target chart template according to the data type of the data set and the query prompt word, and generates a chart template generation statement of the target chart template according to the template parameters, wherein the template parameters include chart type, color scheme, axis labels and legends.

[0014] In addition, to achieve the above purpose, the present application also proposes a large-screen data visualization device based on a large model, and the large-screen data visualization device based on a large model includes: An input module, used to obtain externally input user voice data, perform intent analysis on the user voice data, and determine a query prompt word representing the intent from the user voice data; An analysis module, used for inputting the query prompt word into a preset large model for analysis, and generating target data corresponding to the query prompt word and a target chart template matching the query prompt word; A mapping module maps the target data to the target chart template, determines the target chart, and visualizes the target chart.

[0015] In addition, to achieve the above-mentioned objectives, the present application also proposes a large-screen data visualization device based on a large model, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the large-screen data visualization method based on a large model as described above.

[0016] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by the processor, the steps of the large-screen data visualization method based on the large model as described above are implemented.

[0017] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, it implements the steps of the large-screen data visualization method based on the large model as described above.

[0018] The present application obtains externally input user voice data, performs intent analysis on the user voice data, and determines query prompt words that represent the intent from the user voice data; inputs the query prompt words into a preset large model for analysis, generates target data corresponding to the query prompt words and a target chart template that matches the query prompt words; maps the target data to the target chart template, determines the target chart, and visualizes the target chart. The present application utilizes speech recognition technology and natural language processing technology to interact between users and data, and can effectively and quickly convert user voice commands into intuitive chart displays, and quickly respond to large-screen data visualization commands based on the large model, thereby avoiding the defect that chart parameters need to be reconfigured for visualization when business needs change, and greatly improving the efficiency of data display. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0021] Figure 1A flow chart of the first embodiment of the method for large-screen visualization of data based on a large model provided in this application; Figure 2 A flow chart of the first embodiment of the method for large-screen visualization of data based on a large model provided in this application; Figure 3 A flow chart of the second embodiment of the method for large-screen visualization of data based on a large model provided in this application; Figure 4 This is a schematic diagram of the module structure of a large-screen data visualization device based on a large model according to an embodiment of the present application; Figure 5 This is a schematic diagram of the device structure of the hardware operating environment involved in the large-screen data visualization method based on a large model in an embodiment of the present application.

[0022] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0023] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0024] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0025] It should be noted that the execution subject of this embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions, a large data screen, etc. The following takes a large data screen as an example to illustrate this embodiment and the following embodiments.

[0026] Based on this, the embodiment of the present application provides a data large-screen visualization method based on a large model, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the large-screen data visualization method based on a large model of the present application.

[0027] In this embodiment, the large-screen data visualization method based on the large model includes steps S10 to S30: Step S10, obtaining externally input user voice data, performing intent analysis on the user voice data, and determining a query prompt word representing the intent from the user voice data; It should be noted that user voice data is the user's voice signal captured by a microphone or other audio input device. In various environments, such as conference rooms, homes, offices, etc., users can trigger the data screen to perform specific tasks through verbal commands. Subsequently, the captured user voice data can be converted into text using Automatic Speech Recognition (ASR) technology. Natural language processing (NLP) technology can be used to analyze the text of the converted user voice data to identify the operations the user wants to perform or the content of the inquiry to understand the user's intention. Keywords or phrases are extracted from the text that has undergone intent analysis. These keywords or phrases constitute query prompts. Technologies such as word frequency statistics, TF-IDF (term frequency-inverse document frequency), and named entity recognition (NER) can be used to determine which words can be selected as query prompts.

[0028] It can be understood that through the application of a series of key technologies, the transformation from user voice data to operational query prompts has been achieved, which provides a basis for subsequent data analysis and visualization and improves the intelligence and interactivity of the system.

[0029] In one possible implementation, see Figure 2 , step S10 may include steps A10 to A40: Step A10, converting the user voice data into text data; It should be noted that this step converts the sound signal into text through speech recognition, wherein the main technical means is automatic speech recognition (ASR) technology. The ASR system usually includes multiple components such as sound signal processing, feature extraction, acoustic model, language model, etc., which can convert the user's voice commands into text form that can be further processed.

[0030] Optionally, corresponding to the user voice data, noise removal, volume adjustment, etc. can be performed to improve the accuracy of voice recognition.

[0031] Step A20, performing intent analysis on the text data to determine the user's intent; It should be noted that intent analysis requires the application of natural language processing (NLP) technology. Semantic analysis is a branch of natural language processing that uses semantic analysis technology to understand the meaning of words, phrases, and sentences in text. Intent recognition is to infer the specific content of the user's intended operation or query from text data. Machine learning models can be used to identify the user's intent.

[0032] It can be understood that step A20 implements the process of performing intent analysis on text data to determine the user's intent.

[0033] Step A30, identifying entity keywords in the text data, wherein the entity keywords include one or more of topic keywords, time range keywords, and location range keywords; It should be noted that named entity recognition is a natural language processing task used to identify specific types of entities in text, such as names of people, places, and organizations. Machine learning or deep learning methods are usually used to perform entity recognition. Topic keywords are words related to the main content of the text or the topic of discussion, time range keywords are words that represent time information, such as dates, time periods, etc., and location range keywords are words that represent geographic location information, such as city names, country names, etc.

[0034] Step A40 , based on the user intention and the entity keyword, the entity keyword is converted from colloquial language to a standardized query prompt word that represents the user intention.

[0035] It should be noted that through context understanding, synonym replacement and other methods, colloquial expressions can be converted into standardized expressions to facilitate subsequent query operations. For example, the NLP algorithm can convert the colloquial prompt words of the query into standardized search terms. For example, the "car accident data" input by the user can be converted into "traffic accident data" through semantic similarity search and then search for related data.

[0036] This implementation uses automatic speech recognition (ASR) technology to convert received user voice data into text data in text form, uses natural language processing (NLP) technology to perform intent analysis on text data to understand the actual needs of users, uses named entity recognition (NER) and other related natural language processing technologies to identify key entities in text data, namely entity keywords, and combines user intent and identified entity keywords to construct accurate query prompts. Through the above steps, the system can accurately understand the needs of users and perform corresponding operations based on query prompts, making human-computer interaction more efficient and intuitive.

[0037] Step S20, inputting the query prompt word into the preset large model for analysis, generating target data corresponding to the query prompt word and a target chart template matching the query prompt word; It should be noted that the preset large model is used to analyze query prompts, understand the meaning of query prompts and judge the query intention behind them, and predict the data content and chart content required by users. According to the analysis results of the query prompts, relevant data sets are retrieved from the database or data source, and the appropriate chart template is determined.

[0038] Optionally, the preset large model will analyze the meaning of the query prompt word and determine the specific intention behind it, such as whether to view the trend of sales data, compare the performance of different products, etc., and generate instructions about the data structure, type and source according to the content of the query prompt word. Usually, the database query statement of the target data can be obtained to obtain the required target data from the corresponding database or data warehouse. At the same time, the most appropriate chart template is selected according to the type of data and the intention of the query prompt word, such as line graphs, bar graphs, pie charts, scatter graphs, etc., or diversified chart templates combining multiple chart types. The chart style can be adjusted according to the specific situation, such as color schemes, labels, titles, etc.

[0039] It can be understood that step S20 can achieve the transformation from natural language query prompts to specific data visualization, providing users with an intuitive way to understand and explore data. This automated process greatly simplifies the workflow of data visualization.

[0040] Step S30, mapping the target data to a target chart template, determining the target chart, and visualizing the target chart.

[0041] It should be noted that the mapping process includes mapping the target data generated in step S20 to a chart template that matches the query prompt, and each data point or data series is assigned to a corresponding position in the chart, such as a coordinate axis, color, size, or other visual element. A chart may contain a variety of elements, such as the X-axis and Y-axis: defining the range, scale, and label of the coordinate axis; data: how each data is represented in the chart, such as the color, width, or shape of the line; and titles and comments: adding a title and other comments to the chart to help explain the content of the chart.

[0042] Specifically, you need to determine how to map data points to different parts of the chart, such as the position, color, shape, etc. on the coordinate axis. At the same time, you need to consider different data types and magnitudes to ensure that the chart can adapt to the data display needs in various situations. For the elements in the chart, you need to set the minimum value, maximum value, scale spacing, etc. of the coordinate axis according to the specific target data, define the visual attributes of each data, such as line style, marker shape, etc., add legends to distinguish different data, and provide necessary annotations to explain the key information of the chart. In addition, you can also make adaptive adjustments to the style of the chart, including color schemes, fonts and text, backgrounds and fills.

[0043] It can be understood that step S30 realizes mapping the target data into the chart template, creating and displaying the final chart, and providing users with an intuitive method to understand and explore data, which not only improves the efficiency of data visualization, but also enhances the interaction between users and data.

[0044] This embodiment obtains externally input user voice data, performs intent analysis on the user voice data, and determines query prompt words from the user voice data; inputs the query prompt words into a preset large model for analysis, generates target data corresponding to the query prompt words and a chart template matching the query prompt words; maps the target data to the chart template, determines the target chart, and visualizes the target chart. This embodiment uses voice recognition technology and natural language processing technology to interact between users and data, and can effectively convert user voice commands into intuitive chart displays quickly, and quickly respond to large-screen data visualization commands based on the large model, avoiding the defect that chart parameters need to be reconfigured for visualization when business needs change, and greatly improving the efficiency of data display.

[0045] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can be referred to the above introduction, and will not be repeated in the following. Figure 3 , step S20 may include steps S21 to S22: Step S21, inputting the query prompt word into the preset large model, obtaining the database query statement and the chart template generation statement output by the preset large model that match the query prompt word; It should be noted that based on the understanding of the query prompt words, the system needs to generate one or more database query statements. These query statements are usually SQL statements or specific API calls to extract the required data from the database. In addition to database query statements, the system also needs to generate instructions for creating charts based on the content of the query prompt words. These instructions will guide the chart generation tool on how to organize and present data. Among them, the chart template generation statement needs to specify the type of chart template, or select a preset chart template, and can also specify chart elements such as color schemes, axis labels, and legends.

[0046] Step S22, acquiring target data according to a database query statement, and generating a target chart template according to a chart template generation statement.

[0047] It should be noted that the system needs to establish a connection with the target database, and then use the database query statement generated in step S21 to execute the query. At the same time, the system determines the type of chart (such as line chart, bar chart, etc.) according to the chart template generation statement, and then sets the various parameters of the chart (such as axis labels, color schemes, etc.) according to the configuration information in the chart template generation statement, creates the chart using an appropriate chart generation tool or library, and uses the query results as input data.

[0048] It can be understood that step S22 converts the database query statement and chart template generation statement obtained from step S21 into actual data and charts, converts natural language queries into useful information, and helps users better understand and analyze data by effectively executing database queries and generating intuitive charts.

[0049] This implementation method inputs query prompt words into a preset large model, obtains database query statements and chart template generation statements output by the preset large model, understands and parses natural language text, and converts it into machine-executable operation instructions; obtains target data according to the database query statement, obtains chart templates according to the chart template generation statement, and performs specific tasks according to the aforementioned generated instructions. This implementation method realizes an automated and efficient data acquisition and template selection process, reduces the need to manually write SQL queries and select chart types, and also reduces the error rate, and is suitable for scenarios that require rapid response to user needs.

[0050] In a feasible implementation, step S21 may include steps S211 to S212: Step S211, inputting the query prompt word into the preset large model, obtaining the database query statement output by the preset large model, wherein the preset large model determines the data set matching the query prompt word in the preset database based on the query prompt word, and generates and outputs the database query statement corresponding to the data set; It should be noted that the query hint is an optimized natural language description provided by the user to express the specific information they want to retrieve from the database. For example, "find all products with sales exceeding 1 million in 2023". The preset big model first needs to have the ability to understand natural language to parse the key words and phrases in the query hint and understand the user's query intent. Secondly, it needs to be able to identify the key entities in the query hint, such as time range, amount threshold, product category, etc., or it needs to be able to identify the relationship between entities, such as the relationship between product and sales, the relationship between time range and sales, etc. Based on the parsed key entities and their relationships, the preset big model can build the basic structure of the query, such as table name, field name, and conditional filtering logic, and can also optimize the query statement to improve execution efficiency by adding appropriate index conditions and reducing unnecessary connection operations. Convert the query logic constructed in the above content into a standard database query statement for execution in the database. The database query statement is one or a series of SQL statements used to retrieve data sets that meet the query hint requirements from the preset database.

[0051] As an example, suppose the query prompt word entered is: "Find all products with sales exceeding 1 million in 2023". Then the keywords "2023", "sales exceeding 1 million", and "product" can be identified, the entity "2023" corresponds to the time range, "sales exceeding 1 million" corresponds to the numerical condition, and "product" corresponds to the data table. The relationship between the "sales" field in the relationship product table and the condition "exceeding 1 million" is extracted, as well as the relationship between the time range "2023" and the time field of the product table. According to the above query logic, the query table is determined to be the product table, the query field is sales, and the condition is that sales in 2023 are greater than 1 million. At the same time, as a query logic optimization, an index can be used to speed up the query, such as creating an index on the "sales" field of the product table. Finally, a database query statement can be generated: SELECT * FROM productsWHERE sales>1000000 AND year BETWEEN '2023-01-01' AND '2023-12-31'.

[0052] It is understandable that in this step, natural language processing technology needs to be applied to parse query prompts, understand query intent, identify entities and relationships, and effectively convert query prompts in natural language form into accurate database query statements, thereby achieving effective retrieval of the database.

[0053] Step S212, obtaining preset chart templates, inputting each preset chart template and query prompt word into the preset big model, obtaining the chart template generation sentence output by the preset big model, wherein the preset big model matches each preset chart template based on the query prompt word, determines the preset chart template that matches successfully, and generates the chart template generation sentence of the preset chart template that matches successfully according to the query prompt word.

[0054] It should be noted that the preset chart templates are a series of predefined chart styles and layout options for users to choose from or recommended by the system. Each preset chart template comes with a descriptive text description that explains what type of data display or specific business scenario the chart is suitable for and other information about the template. The preset big model can match the content of the query prompt with the template description of each preset chart template to find the chart template that best meets the query prompt requirements.

[0055] This embodiment can automatically extract information from query prompt words through a large model, generate database query statements and chart template generation statements, and thus achieve efficient data visualization.

[0056] In a feasible implementation, before step S212, the large-screen data visualization method based on the large model may further include steps B10 to B20: Step B10, creating preset chart templates of different chart types; It should be noted that when creating a template, you need to consider different types of charts and their applicable scenarios. Common chart types include but are not limited to: bar charts: suitable for comparing values ​​of different categories; line charts: showing the trend of data changes over time; pie charts: representing the proportion of each part; scatter plots: exploring the relationship between variables; area charts: similar to line charts, but emphasizing the trend of quantity changes; radar charts: used for multi-dimensional comparisons; heat maps: showing the density or intensity distribution of data; maps: visualizing geographic data; bubble charts: combining scatter plots and size ratios to represent the relationship between three variables.

[0057] For each preset chart template, the following parameters should be included: Chart title: briefly describe the content of the chart; Coordinate axis: including axis labels, scales, etc.; Legend: color identification used to distinguish different data series; Data label: specific numerical display on the data point; Color scheme: colors of different data series; Interactive functions: such as hover prompts, zoom, etc. These templates can be easily applied to different types of data sets to quickly generate visual charts.

[0058] Step B20, construct a template description corresponding to each preset chart template, wherein the template description includes applicable data type, creator, creation date, version, function description, classification label and application scenario.

[0059] It should be noted that the description of each preset chart template should include the following elements: Applicable data type: describe what type of data the template is applicable to, such as numeric, categorical, time series, etc. For example, for a bar chart, it may be applicable to displaying comparisons of values ​​of different categories; Creator: record the name of the individual or team who created the template; Creation date: mark the creation date of the template to track version history; Version: indicate the version number of the template to facilitate tracking of modification records; Functional description: introduce the main functions and features of the template in detail, such as "This template supports dynamic data loading and can update chart content in real time"; Category tags: add tags to the template for easy classification and search, for example, [Trend analysis][Comparison][Time series]; Application scenarios: describe the application scenarios for which the template is most suitable, such as reports, monitoring panels, etc. For example, it is applicable to product sales trend analysis reports.

[0060] As an example, suppose we have a chart template called "Product Sales Trends". Here is an example of its template description: Applicable data types: numeric (product sales), time series (date); Created by: Data Visualization Team; Creation date: 2024-08-28; Version: V1.0; Function description: This template is a dynamic line chart that can show the sales trend of products in a specific time period and supports viewing data by day, week, and month; Category tags: [Trend analysis] [Time series] [Sales]; Application scenarios: sales reports, market analysis, business monitoring.

[0061] It is understandable that through detailed records and descriptions, it is easier to find chart templates that meet the needs, and to better understand and use these templates. This not only improves work efficiency, but also enhances the practical value of chart templates.

[0062] This implementation method creates preset chart templates of different chart types and applies them to different types of data sets to quickly generate visual charts; constructs template descriptions corresponding to each preset chart template, wherein the template description includes the applicable data type, creator, creation date, version, function description, classification label, and application scenario. Through these descriptive information, it is easy to find and select the most suitable template to meet specific visualization needs.

[0063] In a feasible implementation, step S212 may include step C10: Step C10, input the query prompt word into the preset big model, and obtain the chart template generation statement generated by the preset big model, wherein the preset big model determines the data type of the data set according to the data set that matches the query prompt word in the preset database, and matches the template description of each preset chart template according to the data type of the data set and the query prompt word, determines the successfully matched chart template, and generates the chart template generation statement of the successfully matched chart template.

[0064] It should be noted that after defining each preset chart template with a template description in step B20, a suitable preset chart template can be matched according to the template description. The large model first finds relevant data sets from the preset database based on the query prompt word, and determines the data types of these data sets (such as numerical type, category type, time series, etc.), and then matches the template description of the preset chart template according to the data type and the query prompt word to select one or more chart templates that best match the query prompt word as the target chart template.

[0065] Optionally, the big model needs to understand the context in the query prompt to determine the type of dataset associated with it. For example, if the query prompt mentions "time series", the big model will look for a dataset of the time series type. The big model will compare the key information in the query prompt with the template description of each preset chart template. Fields such as "applicable data type", "functional description" and "application scenario" in the template description will be used to determine whether the template matches. The big model selects the best chart template based on the degree of match, and may also use a scoring mechanism to quantify the degree of match. The generation statement contains specific instructions for creating a chart, such as the dataset used, the selected chart type, etc., and may also contain some configuration parameters, such as color scheme, label style, etc.

[0066] This embodiment combines the user's query prompt words and the description information of the preset chart template, and uses the ability of the big model to automatically select and generate the most suitable chart template. This automated method greatly simplifies the chart creation process and also ensures that the generated chart meets the user's specific needs.

[0067] In another feasible implementation, before step S20, the large-screen data visualization method based on the large model may further include: Step C10, generating a chart data set based on a preset multimodal language model; Step C20, fine-tuning the preset large model through the chart data set to determine the fine-tuned preset large model; It should be noted that the preset multimodal language model is ChartLlama, which is a multimodal language model specifically for charts. It can extract information from charts and create charts based on data and requirements. It can understand and generate various types of charts, including bar charts, pie charts, line charts, scatter plots, and so on. In this application, a high-quality dataset designed specifically for chart understanding is generated by ChartLlama, and the preset large model is fine-tuned through the generated dataset, which can provide a variety of chart types and tasks, and support the training and evaluation of large models.

[0068] After fine-tuning, the preset large model can acquire the ability to understand the charts based on fine-tuning, and can recognize and understand the parameters of different preset chart templates, such as chart types, color schemes, axis labels, legends, etc.

[0069] Step S212 may further include step C30: Step C30, input each preset chart template and query prompt word into the fine-tuned preset large model, and obtain the chart template generation statement generated by the fine-tuned preset large model, wherein the fine-tuned preset large model determines the data type of the data set according to the data set that matches the query prompt word in the preset database, and matches the template description of each preset chart template according to the data type of the data set and the query prompt word, determines the target chart template that matches successfully, determines the template parameters of the target chart template according to the data type of the data set and the query prompt word, and generates the chart template generation statement of the target chart template according to the template parameters, and the template parameters include chart type, color scheme, axis label and legend.

[0070] It should be noted that, in this embodiment, the preset large model is a large model that has been fine-tuned for a specific domain data set. After fine-tuning, the preset large model can improve the ability to understand charts, and then can identify and understand the template parameters of different preset chart templates. The fine-tuned preset large model determines the data type of the data set based on the data set that matches the query prompt word in the preset database. The data type may include numeric type, category type, time series type, etc. The model matches the template description of the preset chart template based on the data type of the data set and the query prompt word to determine the target chart template that best meets the requirements, and determines the template parameters of the target chart template based on the data type of the data set and the query prompt word, such as chart type, color scheme, axis label, legend. Generate a specific chart template generation statement based on the template parameters determined above.

[0071] As an example, assuming that a query prompt is provided, "Show sales for each quarter in the past five years", the process of this step is as follows: Based on the query prompt, the model determines the time series data contained in the data set; the model searches for the template that best suits the time series data from the preset chart templates, assuming that a line chart template is found; Based on the query prompt, the chart type is determined to be a line chart, the color scheme may be the default color scheme or a user-specified color scheme, the axis labels are quarter (horizontal axis) and sales (vertical axis), and the legend may contain marks for different years. Based on the above information, a chart template is generated to generate a statement: "Create a line chart that shows sales for each quarter in the past five years, with quarters on the horizontal axis and sales on the vertical axis, using the default color scheme." This implementation enhances the ability of the large model to generate charts by introducing additional fine-tuning steps. By pre-generating a high-quality chart dataset and fine-tuning the large model, the model can more accurately understand the user's intention, identify and understand the parameters of different preset chart templates, and generate chart template generation statements that meet the requirements, which not only improves the quality of chart generation, but also makes the entire process more automated and efficient.

[0072] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the large-screen data visualization method based on a large model of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0073] This application also provides a large-screen data visualization device based on a large model, please refer to Figure 4 , the data large-screen visualization device based on the large model includes: An input module 10 is used to obtain externally input user voice data, perform intent analysis on the user voice data, and determine a query prompt word representing the intent from the user voice data; An analysis module 20 is used to input the query prompt word into a preset large model for analysis, and generate target data corresponding to the query prompt word and a target chart template matching the query prompt word; The mapping module 30 is used to map the target data to the target chart template, determine the target chart, and visualize the target chart.

[0074] The data large-screen visualization device based on a large model provided by the present application adopts the data large-screen visualization method based on a large model in the above-mentioned embodiment, which can solve the technical problem that the large screen needs to be reconfigured for visualization due to changes in business needs. Compared with the prior art, the beneficial effects of the data large-screen visualization device based on a large model provided by the present application are the same as the beneficial effects of the data large-screen visualization method based on a large model provided by the above-mentioned embodiment, and the other technical features of the data large-screen visualization device based on a large model are the same as the features disclosed in the above-mentioned embodiment method, which will not be repeated here.

[0075] The present application provides a large-screen data visualization device based on a large model, and the large-screen data visualization device based on a large model includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the large-screen data visualization method based on the large model in the above-mentioned embodiment one.

[0076] Reference below Figure 5, which shows a schematic diagram of the structure of a large-screen data visualization device based on a large model suitable for implementing the embodiment of the present application. The large-screen data visualization device based on a large model in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The large-scale data large-screen visualization device based on the large model shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0077] like Figure 5 As shown, the data large-screen visualization device based on the large model may include a processing device 1001 (such as a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 to the random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the data large-screen visualization device based on the large model are also stored. The processing device 1001, ROM1002 and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the large-scale data visualization device based on the large model to communicate with other devices wirelessly or by wire to exchange data. Although the large-scale data visualization device based on the large model with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0078] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0079] The data large-screen visualization device based on a large model provided by the present application adopts the data large-screen visualization method based on a large model in the above-mentioned embodiment, which can solve the technical problem that the large screen needs to be reconfigured for visualization due to changes in business needs. Compared with the prior art, the beneficial effects of the data large-screen visualization device based on a large model provided by the present application are the same as the beneficial effects of the data large-screen visualization method based on a large model provided by the above-mentioned embodiment, and the other technical features of the data large-screen visualization device based on a large model are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0080] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0081] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0082] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the large-model-based data large-screen visualization method in the above-mentioned embodiment.

[0083] The readable storage medium provided in this application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned large-screen data visualization method based on a large model, and can solve the technical problem that the large screen needs to be reconfigured for visualization due to changes in business requirements. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as the beneficial effects of the large-screen data visualization method based on a large model provided in the above-mentioned embodiment, and will not be repeated here.

[0084] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the large-screen data visualization method based on a large model as described above.

[0085] The computer program product provided by this application can solve the technical problem that the large screen needs to be reconfigured for visualization due to changes in business requirements. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the large-screen data visualization method based on the large model provided in the above embodiment, which will not be repeated here.

[0086] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A large-screen data visualization method based on a large model, characterized in that: The data large-screen visualization method based on the large model includes: Acquire externally input user voice data, perform intent analysis on the user voice data, and determine a query prompt word representing the intent from the user voice data; Input the query prompt word into a preset large model for analysis, and generate target data corresponding to the query prompt word and a target chart template matching the query prompt word; The target data is mapped to the target chart template, a target chart is determined, and the target chart is visualized.

2. The large-screen data visualization method based on a large model as claimed in claim 1, characterized in that: The step of performing intent analysis on the user voice data and determining a query prompt word from the user voice data comprises: Converting the user voice data into text data; Performing intent analysis on the text data to determine user intent; Identifying entity keywords in the text data, wherein the entity keywords include one or more of topic keywords, time range keywords, and location range keywords; According to the user intention and the entity keyword, the entity keyword is converted from colloquial language to a standardized query prompt word representing the user intention.

3. The large-screen data visualization method based on a large model as claimed in claim 1, characterized in that: The step of inputting the query prompt word into a preset large model for analysis, and generating target data corresponding to the query prompt word and a target chart template matching the query prompt word comprises: Input the query prompt word into the preset large model, and obtain the database query statement and the chart template generation statement output by the preset large model that match the query prompt word; The target data is acquired according to the database query statement, and a target chart template is generated according to the chart template generation statement.

4. The large-screen data visualization method based on a large model as claimed in claim 3, characterized in that: The step of inputting the query prompt word into the preset large model and obtaining the database query statement output by the preset large model and the chart template generation statement includes: Input the query prompt word into the preset large model, and obtain a database query statement output by the preset large model, wherein the preset large model determines a data set matching the query prompt word in a preset database based on the query prompt word, and generates and outputs a database query statement corresponding to the data set; Obtain a preset chart template, input each of the preset chart templates and the query prompt word into the preset big model, and obtain a chart template generation statement output by the preset big model, wherein the preset big model matches each of the preset chart templates based on the query prompt word, determines a successfully matched preset chart template, and generates a chart template generation statement for the successfully matched preset chart template according to the query prompt word.

5. The large-screen data visualization method based on a large model as claimed in claim 4, characterized in that: Before the step of inputting each of the preset chart templates and the query prompt words into the preset large model, the step further includes: Create preset chart templates for different chart types; Construct a template description corresponding to each of the preset chart templates, wherein the template description includes applicable data type, creator, creation date, version, function description, classification label and application scenario.

6. The large-screen data visualization method based on a large model as claimed in claim 5, characterized in that: The step of inputting each of the preset chart templates and the query prompt words into the preset large model to obtain the chart template generation statement output by the preset large model includes: The query prompt word is input into the preset big model, and the chart template generation statement generated by the preset big model is obtained, wherein the preset big model determines the data type of the data set according to the data set matching the query prompt word in the preset database, and matches the template description of each preset chart template according to the data type of the data set and the query prompt word, determines the chart template that matches successfully, and generates the chart template generation statement of the chart template that matches successfully.

7. The large-screen data visualization method based on a large model as claimed in claim 5, characterized in that: Before the step of inputting the query prompt word into a preset large model for analysis, the method further includes: Generate a chart dataset based on a preset multimodal language model; Fine-tune the preset large model using the chart data set to determine the fine-tuned preset large model; The step of inputting each of the preset chart templates and the query prompt words into the preset large model to obtain the chart template generation statement output by the preset large model includes: Input each of the preset chart templates and the query prompt word into a fine-tuned preset large model, and obtain a chart template generation statement generated by the fine-tuned preset large model, wherein the fine-tuned preset large model determines the data type of the data set according to the data set matching the query prompt word in a preset database, and matches the template description of each of the preset chart templates according to the data type of the data set and the query prompt word to determine a target chart template that has been successfully matched, determines the template parameters of the target chart template according to the data type of the data set and the query prompt word, and generates a chart template generation statement of the target chart template according to the template parameters, wherein the template parameters include chart type, color scheme, axis labels and legends.

8. A large-screen data visualization device based on a large model, characterized in that: The data large-screen visualization device based on the large model includes: An input module, used to obtain externally input user voice data, perform intent analysis on the user voice data, and determine a query prompt word representing the intent from the user voice data; An analysis module, used for inputting the query prompt word into a preset large model for analysis, and generating target data corresponding to the query prompt word and a target chart template matching the query prompt word; A mapping module maps the target data to the target chart template, determines the target chart, and visualizes the target chart.

9. A large-screen data visualization device based on a large model, characterized in that: The device comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the large-screen data visualization method based on a large model as described in any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the large-screen data visualization method based on a large model are implemented as described in any one of claims 1 to 7.

Images