Large-scale model-based graphical analysis methods, devices, media, equipment, and products

Abstract

A chart analysis method, apparatus, medium, device, and product based on a large model, relating to large model technology and data analysis technology, is disclosed. The method includes: responding to a trigger operation on a dashboard page, determining a chart analysis template corresponding to the trigger operation; first content being associated with executable code for obtaining second content, the executable code being determined by the large model based on user input, the input indicating the analysis logic for a first data chart in at least one data chart; performing chart analysis on the first data chart by executing the executable code to obtain the second content; replacing the first content in the chart analysis template with the second content, and outputting the chart analysis content based on the replaced chart analysis template. By moving the large model processing to the template configuration stage, the large model avoids repetitive processing of the same logic, improving code reusability, chart analysis efficiency, and accuracy.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of large model technology and data analysis, in particular, to a large model-based chart analysis method, device, medium, equipment and product. BACKGROUND

[0002] In data analysis and decision-making, the dashboard of a BI (Business Intelligence) tool presents key business indicators through visualization to help business personnel quickly grasp business dynamics. However, the interpretation and analysis of the BI dashboard rely on manual analysis, which not only consumes time and effort, but also may lead to inconsistent analysis results due to individual experience differences.

[0003] In related technologies, to solve this problem, BI dashboard data and user input content are directly fed to a large model for automatic interpretation to obtain chart analysis results. Each interpretation and analysis needs to re-process the large model, which is time-consuming and inefficient, and the chart analysis results output by the large model each time are inconsistent, making it difficult to control accuracy. SUMMARY

[0004] This summary is provided to introduce a selection of concepts, which are further described below in the detailed description. This summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter.

[0005] In a first aspect, the present disclosure provides a large model-based chart analysis method, comprising:

[0006] In response to a triggering operation on a dashboard page, determining a chart analysis template corresponding to the triggering operation, the dashboard page displaying at least one data chart, the chart analysis template including first content, the first content being associated with executable code for obtaining second content, the executable code being determined by a large model based on input content of a user, the input content being used to indicate analysis logic for a first data chart in the at least one data chart;

[0007] Performing chart analysis on the first data chart by executing the executable code to obtain the second content;

[0008] Replacing the first content in the chart analysis template with the second content, and outputting chart analysis content according to the chart analysis template obtained after replacement.

[0009] In a second aspect, the present disclosure provides a large model-based chart analysis device, comprising:

[0010] A determination module is configured to determine a chart analysis template corresponding to a triggering operation in response to the triggering operation on a dashboard page, the dashboard page displays at least one data chart, the chart analysis template includes first content, the first content is associated with executable code for obtaining second content, the executable code is determined by a large model based on input content of a user, and the input content is used to indicate an analysis logic of a first data chart in the at least one data chart.

[0011] An execution module is configured to perform chart analysis on the first data chart by executing the executable code, and obtain the second content.

[0012] An output module is configured to replace the first content in the chart analysis template with the second content, and output chart analysis content according to the chart analysis template obtained after the replacement.

[0013] In a third aspect, the present disclosure provides a computer readable medium having a computer program stored thereon, the program being executed by a processing device to implement the steps of the method in the first aspect.

[0014] In a fourth aspect, the present disclosure provides an electronic device, comprising:

[0015] A storage device having a computer program stored thereon;

[0016] A processing device configured to execute the computer program in the storage device to implement the steps of the method in the first aspect.

[0017] In a fifth aspect, the present disclosure provides a computer program product comprising a computer program, the computer program being executed by a processor to implement the steps of the method in the first aspect.

[0018] By the technical solution, a chart analysis template including first content can be configured, the first content is associated with executable code used to obtain second content, the executable code is determined by a large model based on an analysis logic of the user for indicating a first data chart in at least one data chart, by responding to a triggering operation on a dashboard page, a chart analysis template corresponding to the triggering operation is determined, and by executing the executable code, the first data chart is analyzed to obtain the second content, and then the first content in the chart analysis template is replaced with the second content, and according to the chart analysis template obtained after the replacement, chart analysis content is output. By this method, the large model processing process is placed in the template configuration stage, and the chart analysis process is solidified as reusable executable code, so that each time the chart analysis is performed, the corresponding executable code is executed, and the replaceable content of the chart analysis template is replaced with the code execution result to obtain the corresponding chart analysis content, avoiding repeated processing of the large model for the same logic, improving the code reuse rate and the chart analysis efficiency, and effectively improving the accuracy and reliability of the chart analysis content.

[0019] Other features and advantages of the present disclosure will be described in detail in the following detailed description section. BRIEF DESCRIPTION OF DRAWINGS

[0020] The above and other features, advantages, and aspects of embodiments of the present disclosure will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings in which:

[0021] Figure 1 is a flow diagram of a large model free interpretation scheme according to an exemplary embodiment of the present disclosure;

[0022] Figure 2 is a flow diagram of a chart analysis method based on a large model according to an exemplary embodiment of the present disclosure;

[0023] Figure 3 is a schematic diagram of a template configuration page according to an exemplary embodiment of the present disclosure;

[0024] Figure 4 is a schematic diagram of a dashboard page according to an exemplary embodiment of the present disclosure;

[0025] Figure 5 is a schematic diagram of an executable code according to an exemplary embodiment of the present disclosure;

[0026] Figure 6 is a schematic diagram of a dashboard page according to an exemplary embodiment of the present disclosure;

[0027] Figure 7 is a template analysis process diagram of static content combined with intelligent paragraphs according to an example embodiment of the present disclosure;

[0028] Figure 8 is a structural block diagram of a large model-based chart analysis apparatus according to an example embodiment of the present disclosure;

[0029] Figure 9 is a structural diagram of an electronic device according to an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0030] Embodiments of the present disclosure will be described more fully hereinafter with reference to the accompanying drawings. While several embodiments of the present disclosure are shown in the drawings, it is understood that the present disclosure can be embodied in various forms and should not be construed as being limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and fully convey the scope of the present disclosure to those skilled in the art. It should be understood that the drawings and embodiments of the present disclosure are only for illustrative purposes and are not intended to limit the scope of protection of the present disclosure.

[0031] It should be understood that each step described in the method embodiments of the present disclosure can be executed in different orders and / or in parallel. In addition, the method embodiments can include additional steps and / or omit the execution of the steps shown. The scope of the present disclosure is not limited in this respect.

[0032] The term “comprising” and variations thereof as used herein are open-ended, that is, “including but not limited to”. The term “based on” is “based, at least in part, on”. The term “one embodiment” means “at least one embodiment”; the term “another embodiment” means “at least one additional embodiment”; the term “some embodiments” means “at least some embodiments”. Related definitions are given below in the description of the embodiments.

[0033] It should be noted that the terms “first”, “second”, and the like used in the present disclosure are only used to distinguish different devices, modules or units, and are not intended to limit the order or interdependence of the functions performed by these devices, modules or units.

[0034] It should be noted that the terms “one”, “multiple” used in the present disclosure are illustrative and not limiting, and those skilled in the art should understand that “one or more” should be understood unless otherwise explicitly stated in the context.

[0035] The names of the messages or information exchanged between the devices in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of the messages or information.

[0036] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0037] For example, upon receiving a user's active request, a prompt message is sent to the user to explicitly inform them that the requested operation will require the acquisition and use of the user's personal information. This allows the user to independently choose whether to provide personal information to the software or hardware, such as the electronic device, application, server, or storage medium performing the operations of this disclosed technical solution, based on the prompt message.

[0038] As an optional but non-limiting implementation, in response to a user's active request, sending a prompt message to the user can be done via a pop-up window, where the prompt message can be presented in text format. Furthermore, the pop-up window can also include a selection control allowing the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0039] It is understood that the above notification and user authorization process are merely illustrative and do not constitute a limitation on the implementation of this disclosure. Other methods that comply with relevant laws and regulations may also be applied to the implementation of this disclosure.

[0040] Meanwhile, it is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0041] In the manual interpretation of BI dashboards, business personnel need to extract data from the dashboards, calculate trends, and write interpretation and analysis reports. This is not only time-consuming and labor-intensive, but also prone to inconsistencies in analysis results due to differences in individual experience. Figure 1 As shown, in related technologies, to solve this problem, BI dashboard data and user input are directly fed into a large model for automated interpretation. The large model autonomously completes data filtering, data analysis, and generates chart analysis results. However, the large model's autonomous interpretation solution still has certain limitations:

[0042] 1. Difficulty in controlling accuracy: This solution relies on a large model to directly analyze dashboard data and user input to generate chart analysis results. However, the input content in natural language format is not sufficiently explicit and lacks a fixed framework constraint. The large model may deviate from the user's pre-set interpretation, generating chart analysis results that do not conform to the business scenario or are incorrect. In addition, due to the limitations of the large model, the output results will also differ for the same input content, further making it difficult to control its accuracy.

[0043] 2. Low Interpretation and Analysis Efficiency: The large model in this solution needs to process a large amount of unfiltered data and user input, and each interpretation requires a full analysis again, resulting in long interpretation times and low efficiency. In addition, for scenarios with strict timeliness requirements, such as daily and weekly reports, this mode is difficult to meet the demand for rapid output, thus delaying business decisions.

[0044] 3. Insufficient standardization of output format: The format of freely generated chart analysis results is relatively arbitrary and cannot be adapted to fixed business templates, such as standardized daily report format, weekly report format, etc. It requires manual secondary editing and adjustment, which is cumbersome and inefficient.

[0045] In view of this, the present disclosure provides a chart analysis method, apparatus, medium, device and product based on a large model to solve the above-mentioned technical problems.

[0046] The embodiments of this disclosure will be further explained below with reference to the accompanying drawings.

[0047] Figure 2 This is a flowchart illustrating a large-model-based graph analysis method according to an exemplary embodiment of this disclosure, with reference to... Figure 2 The chart analysis method may include the following steps:

[0048] S201: In response to a trigger operation on the dashboard page, determine the chart analysis template corresponding to the trigger operation. The dashboard page displays at least one data chart. The chart analysis template includes first content, which is associated with executable code for obtaining second content. The executable code is determined by a large model based on user input, which instructs the analysis logic for the first data chart in the at least one data chart.

[0049] The chart analysis method provided in this embodiment can be applied to business scenarios that require chart analysis and interpretation. The specific application can be determined according to the requirements, and this disclosure does not impose any restrictions on it.

[0050] For example, such as Figure 3 As shown, the intelligent paragraphs that can be replaced in the first content representation template. The aforementioned large model can be a large language model or a multimodal large model, which can be determined based on the user's input requirements, for example, such as... Figure 3 In the intelligent paragraph 1 shown, if the user expresses their analytical logic about the chart through plain text input in natural language form, a large language model can be selected to process the user input, such as... Figure 3 In the intelligent paragraph 2 shown, if the user needs to express the analytical logic of the chart through a combination of text and graphics, a multimodal model can be selected to process the user input. This disclosure does not impose any restrictions on this.

[0051] Of course, such asFigure 3 As shown, this embodiment is also compatible with the above-mentioned large model free interpretation scheme, and provides a configuration function for configuring prompt words for free interpretation, so that users can choose according to their needs. This disclosure does not impose any restrictions on this.

[0052] S202: Perform chart analysis on the first data chart by executing executable code to obtain the second content.

[0053] S203: Replace the first content in the chart analysis template with the second content, and output the chart analysis content based on the chart analysis template obtained after the replacement.

[0054] For example, such as Figure 4 As shown, the dashboard page can output and display the chart analysis content after replacing the smart paragraph, in order to help users understand the data charts on the dashboard page. The specific settings can be configured according to needs, and this disclosure does not impose any restrictions on this.

[0055] By adopting the above method, the large model processing is moved to the template configuration stage, and the chart analysis process is solidified into reusable executable code. In this way, each time a chart analysis is performed, the corresponding executable code is executed, and the replaceable content of the chart analysis template is replaced with the code execution result to obtain the corresponding chart analysis content. This avoids the large model from repeatedly processing the same logic, improves code reusability and chart analysis efficiency, and effectively improves the accuracy and reliability of the chart analysis content.

[0056] In one possible manner, the chart analysis template is configured as follows: in response to a template configuration operation, a template configuration page is displayed, the template configuration page including a first area for configuring template content; in response to a first input operation in the first area, first content associated with executable code is determined based on the first input operation; and the chart analysis template is obtained based at least on the first content associated with executable code.

[0057] For example, it can trigger something like Figure 4 The template configuration control shown displays the template configuration page for the entire dashboard page. It can also display the corresponding template configuration control for each data chart to access the template configuration page for that data chart. A template configuration menu can also be set in the data analysis platform to trigger the display of the platform's template configuration page for analyzing one or more chart templates. The platform's template configuration page can be configured with chart analysis templates for one or more data charts in the dashboard page according to business needs. It can also be configured with chart analysis templates for generating chart analysis reports. The specific settings can be configured according to requirements, and this disclosure does not impose any restrictions on this.

[0058] For example, such as Figure 3As shown, users can configure the content of the chart analysis template in the left-hand area according to their business needs. For example, they can configure the smart paragraphs in the chart analysis template and the executable code associated with the smart paragraphs to obtain the final chart analysis template. This way, each time a chart analysis is performed, the corresponding executable code is executed, and the smart paragraphs in the chart analysis template are replaced with the code execution results to obtain the corresponding chart analysis content. This avoids large models repeatedly processing the same logic, improves code reusability and chart analysis efficiency, and effectively enhances the accuracy and reliability of the chart analysis content.

[0059] It is important to note that, considering that the chart analysis template can be configured with multiple smart paragraphs, a unique identifier can be set for each smart paragraph and its corresponding executable code. For example, smart paragraph 1 corresponds to executable code 1. When replacing the chart analysis template later, smart paragraph 1 will be replaced with the code execution result corresponding to executable code 1, so as to achieve accurate calling of smart paragraphs and accurate backfilling of dynamic content.

[0060] In some possible approaches, the chart analysis method further includes: in response to a fourth input operation in the first area, determining the input content corresponding to the fourth input operation as static content. A chart analysis template is obtained based at least on the first content associated with executable code, including: obtaining the chart analysis template based on the static content and the first content associated with executable code.

[0061] For example, such as Figure 3 As shown, chart analysis templates can include not only replaceable smart paragraphs but also fixed static content. By combining static content with dynamic smart paragraphs to form the skeleton structure of the chart analysis template, the static text and dynamic analysis generation areas are clearly distinguished. This ensures both formatting consistency and retains the dynamic analysis capabilities of the large model, improving the efficiency and accuracy of chart analysis.

[0062] For example, users can set a fixed template skeleton for analytical content based on business needs, such as daily or weekly report templates. The skeleton includes static content such as report titles and fixed introductory descriptions, as well as reserved smart paragraphs—the locations for dynamically generated and populated analytical content. For instance, a sales weekly report template skeleton could include fixed chapter titles such as "This Week's Sales Overview," "Core Indicator Analysis," and "Month-on-Month Trend Interpretation." Then, specific content areas under the "Core Indicator Analysis" chapter could be set as smart paragraphs, the specifics of which can be determined based on the actual business scenario; this disclosure does not impose any restrictions. Through the collaborative design of static content and smart paragraphs, the flexibility and standardization of chart analysis templates are effectively improved, thereby increasing the efficiency and accuracy of chart analysis and ensuring consistent output format and data accuracy.

[0063] In one possible manner, in response to a first input operation in a first region, determining first content associated with executable code based on the first input operation includes: in response to a call configuration operation in the first region, displaying first preset content associated with preset executable code; and in response to a selection operation of third content in the first preset content, determining the first content associated with executable code based on the third content and the preset executable code associated with the third content.

[0064] In this embodiment, preset content associated with executable code can be pre-configured, with different preset content corresponding to different analysis logic. That is, the user's analysis logic for different data charts and analysis dimensions is converted into executable code and corresponding preset content is configured in advance using a large model. For example, preset content 2 corresponds to preset executable code 2. This parameterizes the user's analysis logic in advance, forming preset content associated with preset executable code that the user can select and call. This effectively improves the reusability and maintainability of the executable code. When the analysis logic changes subsequently, the template content does not need to be modified; instead, the preset executable code associated with the preset content can be maintained to meet the user's new analysis needs.

[0065] For example, such as Figure 3 As shown, a predefined calling symbol, such as the "@" symbol, can be used. After the user calls the symbol in the first area, a floating layer can display all available preset content. In response to the user's continued input, the system can dynamically filter out preset content that matches the content entered after the calling symbol. This selected preset content is then used as a smart paragraph in the chart analysis template, automatically associated with corresponding executable code. Alternatively, in response to the user's triggering of the smart paragraph insertion control, all available preset content can be displayed, and the user-selected preset content can be inserted into the current input position of the chart analysis template, resulting in a smart paragraph associated with executable code. The specific choice can be made according to requirements, and this disclosure does not impose any limitations on this.

[0066] Therefore, by providing users with parameterizable preset content, users do not need to repeatedly input the same analysis logic each time they configure a chart analysis template, nor do large models need to repeatedly process the same analysis logic, effectively improving the efficiency of template configuration.

[0067] In one possible manner, the template configuration page further includes a second area for configuring executable code associated with the first content; in response to a first input operation in the first area, determining the first content associated with executable code based on the first input operation includes: in response to a content configuration operation in the first area, determining the first content based on the content corresponding to the content configuration operation, and displaying a configuration input box for the first content in the second area; in response to a second input operation in the configuration input box, generating executable code associated with the first content based on the first input content corresponding to the second input operation through a large model, thereby obtaining the first content associated with executable code.

[0068] For example, such as Figure 3 As shown, after a smart paragraph is entered in the left area, the corresponding configuration input box for that smart paragraph is displayed in the right area. This allows users to describe the analysis logic of the smart paragraph to the large model in natural language. The large model then performs content understanding based on the input and generates executable code that can implement the user's analysis logic. This brings the large model processing to the template configuration stage, eliminating the need for repeated large model analysis during subsequent chart analysis. Simply executing the corresponding executable code will yield the chart analysis content, effectively improving the efficiency and accuracy of chart analysis.

[0069] In one possible manner, in response to a content configuration operation in the first area, determining the first content based on the content corresponding to the content configuration operation includes: in response to a trigger operation of a content insertion control in the first area, displaying second preset content in the first area, and determining the first content based on the second preset content; and / or, in response to a third input operation in the first area, displaying second input content corresponding to the third input operation in the first area, and in response to a content conversion operation of the second input content, converting the second input content into the first content.

[0070] For example, such as Figure 3 As shown, the smart paragraph control can be triggered, displaying default preset content (e.g., "xxxx") at the current input position in the left area, and the corresponding configuration input box in the right area. Furthermore, users can directly use the preset content as the smart paragraph displayed in the template, or modify the preset content as needed. The specific settings can be customized according to requirements, and this disclosure does not impose any restrictions on this.

[0071] For example, such as Figure 3As shown, users can also input static content in the left area, then select the input to trigger a control that converts the selected input content into a smart paragraph, displaying the corresponding configuration input box in the right area. By providing different smart paragraph configuration methods, the flexibility and efficiency of chart analysis template configuration are improved.

[0072] In one possible approach, the first input content includes a first sub-content describing the chart analysis logic in natural language and chart information, the chart information being used to indicate the first data chart and / or data fields in the first data chart. An executable code associated with the first content is generated by a large model based on the first input content corresponding to the second input operation. This includes: performing content understanding on the first sub-content using the large model, and calling the chart data of the first data chart based on the chart information; generating initial code based on the content understanding result and the chart data; and determining the executable code associated with the first content based on the initial code.

[0073] It is important to note that in related technologies, the scope of chart analysis is usually not specified, requiring a global analysis for each chart analysis. However, the chart analysis template of this disclosure, when configuring intelligent paragraphs, allows for input of analysis logic in natural language, as well as chart analysis targeting one or more data charts and specific chart fields. Specifically, the input content can reference specific data charts or chart fields according to business needs, such as "retrieving 7 days' data for indicator xx in the dashboard." Furthermore, it supports complex calculations and decisions, such as "calculating the write-off rate based on www and yyy, rounding to two decimal places; marking × if below 90%, and √ if above 90%", etc. The specific analysis logic can be input according to business requirements.

[0074] For example, such as Figure 4 As shown, the preview control can be triggered to initiate a large model call. The large model generates reusable executable code based on the analysis logic input by the user and references the chart data in the dashboard page. This code can include calculation logic, conditional judgments, etc., resulting in intelligent paragraphs associated with executable code to meet different chart analysis needs of users.

[0075] Furthermore, the generated executable code can be stored as metadata for the smart paragraph, forming the generation logic of that smart paragraph, and, as... Figure 5 As shown, the executable code can be immediately invoked after generation to obtain the result for user preview. Based on the natural language description logic to code conversion mechanism, the user's natural language description analysis logic is converted into executable code, ensuring the accurate solidification and reuse of the analysis logic, simplifying the code generation process, and improving code generation efficiency.

[0076] In addition, continue to refer to Figure 5 It can also switch the display of the analysis logic corresponding to the intelligent paragraph, so that users can modify the analysis logic. It can also display the thought process of the large model in generating executable code based on the analysis logic, so that users can understand the processing logic of the large model.

[0077] In one possible manner, the first input content further includes a second sub-content for describing the content output format. Determining the executable code associated with the first content based on the initial code includes: performing chart analysis on the first data chart by executing the initial code to obtain a fourth content, and displaying the fourth content in the result preview area of ​​the second region according to the content output format required by the second sub-content; if the fourth content meets preset conditions, determining the initial code as the executable code associated with the first content; if the fourth content does not meet preset conditions, in response to a modification operation on the initial code, determining the executable code associated with the first content based on the modified initial code; or, in response to a modification operation on the first input content, generating a new initial code based on the modified first input content through a large model, and determining the executable code associated with the first content based on the new initial code.

[0078] For example, users can also input formatting information to describe the output format, so that the large model can organize the returned results according to the original text format, such as the paragraph structure and wording style selected by the user, or the format redefined in the analysis logic. This satisfies the user's formatting requirements for the input content. Furthermore, by uniquely identifying and associating intelligent paragraphs, executable code, and output format, precise recall of intelligent paragraphs and accurate backfilling of dynamic content can be achieved.

[0079] For example, if the preview results do not meet the user's chart analysis needs, it supports going back to modify the previously entered analysis logic and then generating new executable code, or as follows: Figure 5 As shown, the generated executable code can also be modified directly, and the results can be previewed again after saving. This allows for dynamic adjustment of the executable code based on the preview results, further improving the accuracy of chart analysis.

[0080] It is important to note that the configuration process for smart paragraphs can be executed in parallel. For example, when previewing a smart paragraph, it is relatively time-consuming because it requires calling a large model for processing. At this time, a new smart paragraph can be configured, thereby improving the efficiency of template configuration.

[0081] Furthermore, the configured chart analysis templates can be adjusted, such as by adding or deleting smart paragraphs, and this disclosure does not impose any restrictions on this.

[0082] After the template is configured, when data charts need to be analyzed and interpreted, the stored chart analysis template and the metadata corresponding to each smart paragraph are called. Based on the chart data in the latest dashboard page, the executable code in the metadata of the smart paragraphs is executed to dynamically generate the replacement content of each smart paragraph, complete the assembly of the template framework and the dynamically generated content of the smart paragraphs, and finally output the complete chart analysis content.

[0083] Similarly, with the intelligent paragraph configuration process type, the execution flow of executable code can also be executed concurrently to improve the efficiency of chart analysis.

[0084] In one possible manner, in response to a triggering operation on the dashboard page, a chart analysis template corresponding to the triggering operation is determined, including: in response to a triggering operation on the page entry or analysis control of the dashboard page, a first chart analysis template is determined, which is used to generate first chart analysis content displayed on the dashboard page. The chart analysis content is then output, including: displaying the first chart analysis content in a first preset area or a second preset area associated with the first data chart on the dashboard page.

[0085] For example, such as Figure 4 As shown, when a user enters the dashboard page, they can determine the chart analysis template for global analysis or for a single data chart, and then generate the corresponding chart analysis content. The chart analysis content for global analysis can be displayed at the top of the dashboard page, while the chart analysis content for a single data chart can be displayed in or around the display area of ​​the data chart or in a floating layer. The specific selection can be made according to the needs, and this disclosure does not impose any restrictions on it.

[0086] For example, the chart analysis content of the global analysis can be displayed by default. Then, for individual charts, the user needs to trigger the corresponding analysis control to determine the corresponding chart analysis template, and then display the chart analysis results based on the template. Alternatively, the chart analysis content can be not displayed by default. In response to the user's trigger operation on the analysis control, the corresponding chart analysis template can be determined and the corresponding chart analysis content can be displayed.

[0087] It's important to note that the global analysis here can be applied to all data charts on the dashboard page, or it can be applied to one or more specific data charts. The choice depends on the data charts specified by the user when configuring the template; this disclosure does not impose any restrictions. This allows us to meet users' chart analysis needs for different data ranges.

[0088] In one possible manner, in response to a triggering operation on the dashboard page, a chart analysis template corresponding to the triggering operation is determined, including: in response to a triggering operation on a report generation control on the dashboard page, a second chart analysis template is determined, which is used to generate second chart analysis content in report form. The chart analysis content is then output, including: outputting the second chart analysis content in report form.

[0089] For example, the chart analysis content of the global analysis can be directly exported as a chart analysis report, or... Figure 6 As shown, corresponding report templates can be pre-set for different reporting needs, such as daily report templates and weekly report templates. The specific process is the same as the configuration process of the chart analysis template mentioned above. Then, after triggering the report generation control on the dashboard page, the corresponding options will be displayed. For example, if the user selects "Generate Daily Report", the daily report template will be determined, and a chart analysis report will be generated based on the daily report template, thereby improving the generation efficiency and flexibility of chart analysis reports.

[0090] In other possible implementations, the chart analysis method further includes: in response to a page configuration operation on a dashboard page, displaying a dashboard configuration page, the dashboard configuration page displaying preset components, the preset components being associated with a third chart analysis template; in response to a selection operation on a first component among the preset components, displaying the first component on the dashboard page, the first component displaying third chart analysis content determined based on the third chart analysis template.

[0091] For example, chart analysis templates can be componentized and configured for display as visualization components on dashboard pages. Figure 6 The core metrics shown can be configured with component display content templates using a combination of static content and intelligent paragraphs. When the page is displayed, the executable code associated with the intelligent paragraphs is executed, and the code execution results are then fed back into the template, resulting in the final component display content shown on the dashboard page. Componentizing the display of chart analysis content effectively improves the richness and readability of the dashboard page's content.

[0092] like Figure 7 As shown, this embodiment provides a large-scale model analysis scheme for dashboard pages based on a combination of templates and intelligent paragraphs. By solidifying the framework of the analysis template and reserving slots for dynamic generation (intelligent paragraphs), the large model and user-inputted analysis logic are pre-solidified into executable code. Then, in the chart analysis stage, the corresponding chart data is called to execute the executable code, and the execution results are assembled back into the analysis template to obtain complete chart analysis content, achieving fast, efficient, accurate and standardized analysis output.

[0093] During the template configuration phase, flexible template settings and logic specifications allow for adaptation to the chart analysis and interpretation needs of various business scenarios, meeting the report format and content depth requirements of different users. Based on a template framework combining static content and intelligent paragraphs, the analysis logic of user-input intelligent paragraphs is pre-defined as executable code through a large model. The intelligent paragraphs and their corresponding executable code are then linked and stored with the template skeleton, supporting operations such as adding, adjusting, and deleting, thus enhancing the scalability and usability of the chart analysis template. Furthermore, the output chart analysis content adheres to the user-defined template framework and format, requiring no secondary editing or adjustments.

[0094] During the chart analysis phase, the system can respond to different user chart analysis operations by selecting the corresponding chart analysis template. During the analysis process, the executable code associated with the chart analysis template is executed directly, and the execution results are reassembled back into the template to obtain complete chart analysis content. This avoids repetitive analysis and calculation of the same logic in large models, significantly shortening the analysis and interpretation time. The analysis objectives and processes are clear, greatly improving the accuracy of the analysis results. This better meets the time-sensitive needs of scenarios such as daily and weekly reports, ensuring that business personnel can obtain interpretation results promptly and carry out subsequent work. It effectively guarantees the quality and efficiency of data chart analysis and interpretation, significantly enhancing the intelligent interpretation capabilities of BI tools, enabling them to better adapt to diverse business scenarios, and effectively reducing the risk of decision-making errors due to interpretation biases.

[0095] Based on the same concept, embodiments of this disclosure also provide a chart analysis device based on a large model, such as... Figure 8 As shown, the chart analysis device 800 includes:

[0096] The determination module 801 is used to determine the chart analysis template corresponding to the trigger operation in response to a trigger operation on the dashboard page. The dashboard page displays at least one data chart. The chart analysis template includes first content, and the first content is associated with executable code for obtaining second content. The executable code is determined by a large model based on user input content. The input content is used to indicate the analysis logic for the first data chart among the at least one data chart.

[0097] Execution module 802 is used to perform chart analysis on the first data chart by executing the executable code to obtain the second content;

[0098] The output module 803 is used to replace the first content in the chart analysis template with the second content, and output the chart analysis content according to the chart analysis template obtained after the replacement.

[0099] Optionally, the chart analysis device 800 further includes a configuration module, the configuration module comprising:

[0100] The display module is used to display a template configuration page in response to a template configuration operation. The template configuration page includes a first area for configuring template content.

[0101] A first determining submodule is configured to, in response to a first input operation in the first region, determine, based on the first input operation, first content associated with the executable code;

[0102] The second determining submodule is used to obtain the chart analysis template based at least on the first content associated with the executable code.

[0103] Optionally, the first determining submodule is used to:

[0104] In response to a configuration call in the first region, first preset content associated with preset executable code is displayed;

[0105] In response to the selection operation of the third content in the first preset content, the first content associated with the executable code is determined based on the third content and the preset executable code associated with the third content.

[0106] Optionally, the template configuration page further includes a second area for configuring the executable code associated with the first content; the first determining submodule is used for:

[0107] In response to a content configuration operation in the first area, the first content is determined based on the content corresponding to the content configuration operation, and a configuration input box for the first content is displayed in the second area;

[0108] In response to a second input operation in the configuration input box, the large model generates executable code associated with the first content based on the first input content corresponding to the second input operation, thereby obtaining first content associated with the executable code.

[0109] Optionally, the first determining submodule is used to:

[0110] In response to a trigger operation that inserts a control into the content of the first area, second preset content is displayed in the first area, and the first content is determined based on the second preset content; and / or,

[0111] In response to a third input operation in the first area, a second input content corresponding to the third input operation is displayed in the first area, and in response to a content conversion operation of the second input content, the second input content is converted into the first content.

[0112] Optionally, the first input content includes a first sub-content describing the chart analysis logic in natural language and chart information, wherein the chart information is used to indicate the first data chart and / or data fields in the first data chart, and the first determining sub-module is used to:

[0113] The first sub-content is understood through the large model, and the chart data of the first data chart is called based on the chart information. Initial code is generated based on the result of the content understanding and the chart data.

[0114] The executable code associated with the first content is determined based on the initial code.

[0115] Optionally, the first input content further includes a second sub-content for describing the content output format, and the first determining sub-module is used for:

[0116] The fourth content is obtained by performing chart analysis on the first data chart by executing the initial code, and the fourth content is displayed in the result preview area of ​​the second area according to the content output format required by the second sub-content.

[0117] If the fourth content meets the preset conditions, the initial code will be determined as the executable code associated with the first content;

[0118] If the fourth content does not meet the preset condition, in response to the modification operation of the initial code, the executable code associated with the first content is determined based on the modified initial code; or, in response to the modification operation of the first input content, a new initial code is generated by the large model based on the modified first input content, and the executable code associated with the first content is determined based on the new initial code.

[0119] Optionally, the chart analysis device 800 further includes:

[0120] The third determining submodule is used to determine the input content corresponding to the fourth input operation as static content in response to the fourth input operation in the first area;

[0121] The second determining submodule is used for:

[0122] The chart analysis template is obtained based on the static content and the first content associated with the executable code.

[0123] Optionally, the determining module 801 is used to:

[0124] In response to a page entry or analysis control trigger operation on the dashboard page, a first chart analysis template is determined, which is used to generate the first chart analysis content displayed on the dashboard page;

[0125] The output module 803 is used for:

[0126] The analysis content of the first chart is displayed in the first preset area or the second preset area associated with the first data chart on the dashboard page.

[0127] Optionally, the determining module 801 is used to:

[0128] In response to a trigger operation on the report generation control in the dashboard page, a second chart analysis template is determined, which is used to generate second chart analysis content in the form of a report;

[0129] The output module 803 is used for:

[0130] Output the analysis content of the second chart in the form of a report.

[0131] Based on the same concept, embodiments of this disclosure also provide a computer-readable medium having a computer program stored thereon, which, when executed by a processing device, implements the steps of any of the above-described large-model-based graph analysis methods.

[0132] Based on the same concept, this disclosure also provides an electronic device that may include:

[0133] A storage device on which computer programs are stored;

[0134] A processing device for executing a computer program stored in a storage device to implement the steps of any of the above-described large-model-based graph analysis methods.

[0135] Based on the same concept, embodiments of this disclosure also provide a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described large-model-based graph analysis methods.

[0136] The following is for reference. Figure 9 The diagram illustrates a structural schematic of an electronic device 900 suitable for implementing embodiments of the present disclosure. The terminal devices in the embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 9The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0137] like Figure 9 As shown, electronic device 900 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 901, which can perform various appropriate actions and processes according to a program stored in read-only memory (ROM) 902 or a program loaded from storage device 908 into random access memory (RAM) 903. RAM 903 also stores various programs and data required for the operation of electronic device 900. Processing device 901, ROM 902, and RAM 903 are interconnected via bus 904. Input / output (I / O) interface 905 is also connected to bus 904.

[0138] Typically, the following devices can be connected to I / O interface 905: input devices 906 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 907 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 908 including, for example, magnetic tapes, hard disks, etc.; and communication devices 909. Communication device 909 allows electronic device 900 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 9 An electronic device 900 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0139] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 909, or installed from a storage device 908, or installed from a ROM 902. When the computer program is executed by a processing device 901, it performs the functions defined in the methods of embodiments of this disclosure.

[0140] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0141] In some implementations, communication can be conducted using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol), and can be interconnected with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0142] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0143] The aforementioned computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: in response to a trigger operation on a dashboard page, determine a chart analysis template corresponding to the trigger operation, the dashboard page displaying at least one data chart, the chart analysis template including first content, the first content being associated with executable code for obtaining second content, the executable code being determined by a large model based on user input, the input being used to instruct analysis logic for the first data chart among the at least one data chart; perform chart analysis on the first data chart by executing the executable code to obtain the second content; replace the first content in the chart analysis template with the second content, and output chart analysis content based on the chart analysis template obtained after the replacement.

[0144] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0145] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0146] The modules described in the embodiments of this disclosure can be implemented in software or hardware. The names of the modules are not, in some cases, intended to limit the functionality of the module itself.

[0147] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0148] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0149] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0150] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0151] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative forms of implementing the claims. Regarding the apparatus in the above embodiments, the specific manner in which the various modules perform their operations has been described in detail in the embodiments relating to the method, and will not be elaborated upon here.

Claims

1. A large model-based chart analysis method, characterized by, The chart analysis method comprises: In response to a triggering operation on a dashboard page, a chart analysis template corresponding to the triggering operation is determined, the dashboard page displaying at least one data chart; wherein the chart analysis template comprises first content, the first content being associated with executable code for obtaining second content, the chart analysis template being configured in the following manner: in response to a template configuration operation, a template configuration page is displayed, the template configuration page comprising a first area for configuring template content and a second area for configuring executable code associated with the first content; in response to a content configuration operation in the first area, the first content is determined based on content corresponding to the content configuration operation, and a configuration input box of the first content is displayed in the second area; in response to a second input operation in the configuration input box, first input content corresponding to the second input operation is obtained, the first input content comprising first sub-content describing chart analysis logic in natural language form and chart information for indicating a first data chart and / or data fields in the first data chart; the first sub-content is subjected to content understanding by a large model, and chart data of the first data chart is called based on the chart information, initial code is generated based on a result of the content understanding and the chart data, and executable code associated with the first content is determined based on the initial code, the chart analysis template being obtained based on the first content and the executable code associated with the first content; The first data chart is subjected to chart analysis by executing the executable code, and the second content is obtained; The first content in the chart analysis template is replaced with the second content, and chart analysis content is output according to the chart analysis template obtained after replacement. 2.The large model-based graph analysis method of claim 1, wherein, The method further comprises: In response to a call configuration operation in the first area, first preset content associated with preset executable code is displayed; In response to a selection operation on third content in the first preset content, first content associated with the executable code is determined based on the third content and preset executable code associated with the third content. 3.The large model-based graph analysis method of claim 1, wherein, The response to the content configuration operation in the first area, the first content being determined based on content corresponding to the content configuration operation, comprises: In response to a triggering operation on a content insertion control in the first area, second preset content is displayed in the first area, and the first content is determined based on the second preset content; and / or, In response to a third input operation in the first area, second input content corresponding to the third input operation is displayed in the first area, and in response to a content conversion operation on the second input content, the second input content is converted into the first content.

4. The large model-based graph analysis method according to any one of claims 1-3, characterized in that, The first input content further comprises second sub-content for describing content output format, and the determination of the executable code associated with the first content based on the initial code comprises: The first data chart is analyzed by executing the initial code to obtain fourth content, and the fourth content is displayed in a result preview area of the second area in a content output format required by the second sub-content; In a case where the fourth content meets a preset condition, the initial code is determined as executable code associated with the first content; In a case where the fourth content does not meet the preset condition, in response to a modification operation on the initial code, executable code associated with the first content is determined based on the modified initial code, or, in response to a modification operation on the first input content, new initial code is generated based on the modified first input content by using the large model, and executable code associated with the first content is determined based on the new initial code.

5. The large model-based graph analysis method according to any one of claims 1-3, characterized in that, The chart analysis method further includes: In response to a fourth input operation in the first area, input content corresponding to the fourth input operation is determined as static content; The at least based on the first content associated with the executable code to obtain the chart analysis template includes: Based on the static content and the first content associated with the executable code, the chart analysis template is obtained.

6. The large model-based graph analysis method according to any one of claims 1-3, characterized in that, The response to the triggering operation on the dashboard page to determine the chart analysis template corresponding to the triggering operation includes: In response to a triggering operation on a page entry or an analysis control of the dashboard page, a first chart analysis template is determined, and the first chart analysis template is used to generate first chart analysis content displayed on the dashboard page; The output chart analysis content includes: The first chart analysis content is output and displayed in a first preset area in the dashboard page or a second preset area associated with the first data chart.

7. The large model-based graph analysis method according to any one of claims 1-3, characterized in that, The response to the triggering operation on the dashboard page to determine the chart analysis template corresponding to the triggering operation includes: In response to a triggering operation on a report generation control in the dashboard page, a second chart analysis template is determined, and the second chart analysis template is used to generate second chart analysis content in the form of a report; The output chart analysis content includes: The second chart analysis content is output in the form of a report.

8. A large model-based graph analysis apparatus, characterized by, The chart analysis apparatus includes: The determining module is configured to determine a chart analysis template corresponding to a triggering operation in response to the triggering operation on a dashboard page, the dashboard page displaying at least one data chart; wherein the chart analysis template comprises first content, the first content being associated with executable code for obtaining second content, and the chart analysis template being configured in the following manner: in response to a template configuration operation, displaying a template configuration page, the template configuration page comprising a first area for configuring template content and a second area for configuring executable code associated with the first content; in response to a content configuration operation in the first area, determining the first content based on the content configuration operation, and displaying a configuration input box of the first content in the second area; in response to a second input operation in the configuration input box, obtaining first input content corresponding to the second input operation, the first input content comprising first sub-content describing chart analysis logic in natural language and chart information for indicating a first data chart and / or a data field in the first data chart; performing content understanding on the first sub-content by a large model, calling chart data of the first data chart based on the chart information, generating initial code based on a result of the content understanding and the chart data, and determining executable code associated with the first content based on the initial code, and obtaining the chart analysis template based on the first content and the executable code associated with the first content; The executing module is configured to perform chart analysis on the first data chart by executing the executable code, and obtain the second content. The output module is configured to replace the first content in the chart analysis template with the second content, and output chart analysis content according to the chart analysis template obtained after the replacement.

9. A computer readable medium having stored thereon a computer program, characterized in that, The computer program is executed by the processing device to implement the steps of the method of any one of claims 1-7.

10. An electronic device, comprising: The computer program is executed by the processing device to implement the steps of the method of any one of claims 1-7. The computer program is executed by the processing device to implement the steps of the method of any one of claims 1-7. ​ 11. A computer program product comprising a computer program, characterized in that, ​

Images