Method and device for generating data analysis large screen and data analysis system

Abstract

The invention provides a data analysis large screen generation method and device and a data analysis system.The method comprises the steps that an input data analysis requirement text is obtained, and the data analysis requirement text is a text describing a data analysis requirement; the data analysis demand text, the chart type and the data field are matched, a target chart type and a target data field are obtained, the target chart type is the chart type meeting the data analysis demand, and the target data field is the name of data needed by meeting the data analysis demand; matching a chart template according to the target chart type to obtain a target template; acquiring data corresponding to the target data field and filling the data into the target template to obtain a data chart; and generating a data analysis large screen according to the data chart. The problem of low efficiency of data analysis in the prior art is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data analysis, in particular to a data analysis big screen generation method and device, a computer readable storage medium and a data analysis system. BACKGROUND

[0002] The current mainstream data analysis platform has significant technical limitations: first, traditional systems (such as Tableau, Power BI) highly rely on manual modeling processes, and face dual challenges of efficiency and accuracy when dealing with high-dimensional data. Second, traditional BI requires a large amount of data processing and developer input. SUMMARY

[0003] The main purpose of the present application is to provide a data analysis big screen generation method, device, computer readable storage medium and data analysis system to at least solve the problem of low efficiency of data analysis in the prior art.

[0004] In order to achieve the above purpose, according to one aspect of the present application, a data analysis big screen generation method is provided, comprising: obtaining an input data analysis requirement text, the data analysis requirement text being a text describing data analysis requirements; matching the data analysis requirement text, a chart type and a data field to obtain a target chart type and a target data field, the target chart type being the chart type that meets the data analysis requirements, and the target data field being the name of the data required to meet the data analysis requirements; matching a chart template according to the target chart type to obtain a target template; obtaining data corresponding to the target data field and filling the target template to obtain a data chart; and generating a data analysis big screen according to the data chart.

[0005] Optionally, matching the data analysis requirement text, the chart type and the data field to obtain the target chart type and the target data field comprises: associating text semantics with the data field through a contrast learning technology to obtain a first mapping relationship; associating the text semantics with the target chart type through the contrast learning technology to obtain a second mapping relationship; parsing the data analysis requirement text to obtain target text semantics; querying the data field corresponding to the target text semantics according to the first mapping relationship to obtain the target data field; determining the chart type corresponding to the field type of the target data field as the target chart type, or querying the chart type corresponding to the target text semantics according to the second mapping relationship to obtain the target chart type, the field type including a text type field and a numerical type field.

[0006] Optionally, after determining the chart type corresponding to the field type of the target data field as the target chart type, or querying the chart type corresponding to the target text semantics according to the second mapping relationship to obtain the target chart type, the method further comprises: determining the chart type corresponding to the field type of the target data field as a first target chart type; querying the chart type corresponding to the target text semantics according to the second mapping relationship to obtain a second target chart type; in the case that the first target chart type and the second target chart type are inconsistent, adjusting the first mapping relationship and / or the second mapping relationship so that the first target chart type and the second target chart type are consistent.

[0007] Optionally, matching the data analysis requirement text, chart type and data field to obtain a target chart type and a target data field comprises: constructing a knowledge graph, wherein the nodes of the knowledge graph comprise text semantics, the chart type and the data field, and the edges of the knowledge graph are semantic associations between the nodes; parsing the data analysis requirement text to obtain a target text semantics; querying the knowledge graph according to the target text semantics to obtain the target chart type and the target data field.

[0008] Optionally, matching a chart template according to the target chart type to obtain a target template comprises: matching the target chart type with a layout structure of a template to obtain the target template, wherein the chart types corresponding to the layout structure comprise the chart types corresponding to core KPIs, the chart types corresponding to spatial distribution and the chart types corresponding to time series trends.

[0009] Optionally, obtaining data corresponding to the target data field and filling the target template to obtain a data chart comprises: obtaining data corresponding to the target data field and filling the target template to obtain an initial data chart; using AI to analyze the initial data chart to obtain an index transformation trend and a data chart adjustment strategy, wherein the index transformation trend is a transformation trend of an index corresponding to data of the data chart, and the data chart adjustment strategy is a chart type transformation strategy and a simplified data structure strategy; adjusting the initial data chart according to the data chart adjustment strategy to obtain an adjusted data chart; and adding the index transformation trend to the corresponding adjusted data chart to obtain the data chart.

[0010] Optionally, generating a data analysis large screen according to the data chart comprises: using a dynamic layout algorithm based on an attention mechanism to calculate a visual importance weight of the data chart; setting a position of the data chart according to the visual importance weight to obtain the data analysis large screen, so that the data chart with a visual importance weight greater than a predetermined weight threshold is located in a visual focal point area.

[0011] According to another aspect of the present application, there is provided a data analysis large screen generation apparatus, comprising: an acquisition unit configured to acquire an input data analysis requirement text, the data analysis requirement text being a text describing a data analysis requirement; a first matching unit configured to match the data analysis requirement text, a chart type and a data field to obtain a target chart type and a target data field, the target chart type being the chart type meeting the data analysis requirement, and the target data field being the name of data required to meet the data analysis requirement; a second matching unit configured to match a chart template according to the target chart type to obtain a target template; a filling unit configured to acquire data corresponding to the target data field and fill the target template to obtain a data chart; and a generation unit configured to generate a data analysis large screen according to the data chart.

[0012] According to still another aspect of the present application, there is provided a computer readable storage medium comprising a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform any of the data analysis large screen generation methods.

[0013] According to yet another aspect of the present application, there is provided a data analysis system, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs comprise a program for performing any of the data analysis large screen generation methods.

[0014] By applying the technical solution of the present application, in the data analysis large screen generation method, the data analysis requirement text input by a user is matched with a chart type and a data field to obtain a chart type meeting the data analysis requirement and required data, thereby generating a data chart, and the data chart is laid out and displayed on a large screen to obtain a data analysis large screen. Without developing a corresponding data analysis large screen generation script for different data analysis requirements, the work load of repeated development of generating charts is greatly reduced, and the efficiency of data analysis is improved. BRIEF DESCRIPTION OF DRAWINGS

[0015] The accompanying drawings, which form a part of the present description, are included to provide a further understanding of the application and are incorporated in and constitute a part of this application. The embodiments of the present application, and their

[0016] Figure 1 Fig. 1 shows a hardware structure block diagram of a mobile terminal for performing a data analysis large screen generation method according to an embodiment of the present application;

[0017] Figure 2 Fig. 1 shows a flow diagram of a method for generating a data analysis large screen according to an embodiment of the present application;

[0018] Figure 3 Fig. 2 shows a block diagram of a device for generating a data analysis large screen according to an embodiment of the present application.

[0019] Among the above drawings, the following reference signs are included:

[0020] 102, processor; 104, memory; 106, transmission device; 108, input / output device. DETAILED DESCRIPTION

[0021] It should be noted that the embodiments and features of the present application can be combined with each other in the case of no conflict. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0022] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0023] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those clearly listed steps or units, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0024] As introduced in the background, the efficiency of data analysis in the prior art is low. To solve this problem, the embodiments of the present application provide a method and device for generating a data analysis large screen, a computer readable storage medium and a data analysis system.

[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings.

[0026] The methods and embodiments provided in this application can be executed on a mobile terminal, computer terminal, or similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a data analysis large screen generation method according to an embodiment of the present invention. For example... Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0027] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the data analysis large screen generation method in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0028] A method for generating a data analysis large screen running on a mobile terminal, a computer terminal or the like computing device is provided in the embodiment. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown.

[0029] Figure 2 is a flowchart of the method for generating a data analysis large screen according to the embodiment of the present application. As shown in Figure 2 , the method comprises the following steps:

[0030] Step S201, obtaining an input data analysis requirement text, the data analysis requirement text being a text describing a data analysis requirement;

[0031] Specifically, the user inputs a target analysis theme, i.e. the data analysis requirement text, through an interactive pop-up window, such as "inventory management and product distribution analysis".

[0032] Step S202, matching the data analysis requirement text, a chart type and a data field to obtain a target chart type and a target data field, the target chart type being a chart type meeting the data analysis requirement, and the target data field being a name of data required to meet the data analysis requirement;

[0033] Specifically, through multi-modal joint training and contrast learning technology, the text semantics, chart type features and data field attributes input by the user are deeply associated to form a bidirectional semantic mapping of "text→chart type→data field", thereby realizing more accurate chart recommendation and data association.

[0034] Step S203, matching a chart template according to the target chart type to obtain a target template;

[0035] Specifically, the most suitable layout structure is matched from a template library, and the layout structure of the target template corresponds to a chart type including the target chart type.

[0036] Step S204, obtaining data corresponding to the target data field and filling the target template with the data to obtain a data chart;

[0037] Specifically, the data corresponding to the target data field is called to fill the target template, and the data chart is obtained.

[0038] Step S205, generating a data analysis large screen according to the data chart.

[0039] Specifically, the data chart is laid out and displayed on the large screen to obtain the data analysis large screen.

[0040] In the method for generating the data analysis large screen, the data analysis requirement text input by a user is matched with a chart type and a data field to obtain a chart type meeting the data analysis requirement and required data, so as to generate a data chart, and the data chart is laid out and displayed on a large screen to obtain the data analysis large screen. The method does not need to develop a corresponding data analysis large screen generation script for different data analysis requirements, greatly reduces the repeated development workload of generating charts, improves the data analysis efficiency, and solves the problem of low data analysis efficiency in the prior art.

[0041] To improve the accuracy of topic resolution and the adaptability to business scenarios, in an optional implementation, the step S202 includes:

[0042] In step S2021, a text semantic is semantically associated with the data field by a contrastive learning technology to obtain a first mapping relationship.

[0043] In step S2022, the text semantic is semantically associated with the target chart type by the contrastive learning technology to obtain a second mapping relationship.

[0044] In step S2023, the data analysis requirement text is parsed to obtain a target text semantic.

[0045] In step S2024, the target text semantic is used to query the data field corresponding to the target text semantic according to the first mapping relationship to obtain the target data field.

[0046] In step S2025, the chart type corresponding to the field type of the target data field is determined as the target chart type, or the chart type corresponding to the target text semantic is queried according to the second mapping relationship to obtain the target chart type. The field type includes a text type field and a numerical type field.

[0047] In the above embodiments, in the large-screen theme intelligent analysis link, the system realizes the cross-modal semantic alignment of user input text and chart type, data field by constructing a text-chart-data triple-modal joint training framework, breaking through the limitations of traditional NLP model one-way text understanding, and significantly improving the accuracy of theme analysis and the adaptability of business scenarios. The core of the framework is to associate the user input text semantics, chart type characteristics and data field attributes through multi-modal joint training and contrastive learning technology, forming a "text→chart type→data field" bidirectional semantic mapping, so as to realize more accurate chart recommendation and data association. Specifically, the system first introduces a CLIP-like multi-modal alignment model, which aligns the user input text (such as "inventory management") and chart type (such as column chart, heat map) through contrastive learning technology. For example, when the user inputs "inventory management", the model analyzes the core entities in the text (such as "inventory" "product distribution") and analyzes the intent (such as "monitoring" "prediction"), and through the contrastive learning technology, the text semantics and chart type are semantically associated. For example, "inventory management" may be associated with "line chart" (for displaying time trend) and "heat map" (for displaying regional distribution) at the same time, rather than a single chart type. This cross-modal alignment not only relies on the direct mapping of text and chart, but also ensures the semantic consistency of text and chart through similarity calculation in the multi-modal embedding space. In the semantic deconstruction stage, the system further optimizes the semantic association of text and chart type through contrastive learning. The core idea of contrastive learning is to maximize the similarity of positive sample pairs and minimize the similarity of negative sample pairs. For example, in the positive sample pair of "inventory management" and "line chart", the model learns the semantic similarity of text and chart; while in the negative sample pair of "inventory management" and "pie chart", the model reduces the similarity of text and chart. Through this training, the system can dynamically identify the implicit chart type demand in the text and generate more suitable recommendation results for business scenarios. For example, when the user inputs "inventory turnover rate analysis", the system will combine "turnover rate" (time trend) and "analysis" (monitoring demand) in the text to automatically recommend "line chart" (to display time series changes) and "heat map" (to display regional distribution differences), rather than a single chart type, so as to meet the needs of complex business scenarios. In addition, the system further combines the output of the multi-modal model in the knowledge graph mapping stage to dynamically associate the fields in the enterprise data asset directory (such as "inventory quantity" "product category"). For example, when the system identifies "inventory quantity" as a numerical field, it will automatically recommend "column chart" or "line chart" as the adaptive chart type; while identifying "product category" as a text field, it will recommend "pie chart" or "tree chart" as the adaptive chart type.This dynamic association mechanism not only relies on the rule matching of field types and chart types, but also through the semantic understanding ability of the multi-modal model, the data fields are semantically associated with the text topics at a deeper level. For example, when the user inputs "inventory management and product distribution analysis", the system will combine the "inventory management" and "product distribution" two text topics, dynamically map the "inventory quantity" field to the "column chart", and associate the "product category" field to the "pie chart", thereby generating a structured chart configuration scheme.

[0048] In addition, the technology also has dynamic expansion capability. When the enterprise adds new data fields (such as "inventory warning threshold"), the system can automatically expand the chart type recommendation range (such as adding "warning heat map") through the semantic understanding ability of the multi-modal model, without manual intervention. This flexibility enables the system to quickly adapt to new business needs, further improving the intelligent level of data analysis.

[0049] In summary, the large-screen theme intelligent analysis link realizes the deep semantic association of text, chart and data through the triple modal joint training framework and contrast learning technology, not only breaking through the limitations of traditional NLP models, but also significantly improving the accuracy of theme analysis and the adaptability of business scenarios, providing a more intelligent and flexible user interaction experience for data analysis platforms.

[0050] In order to further improve the accuracy of theme analysis, in an optional implementation, after determining the chart type corresponding to the field type of the target data field as the target chart type, or querying the chart type corresponding to the target text semantic according to the second mapping relationship to obtain the target chart type, the method further includes:

[0051] Step S301, determining the chart type corresponding to the field type of the target data field as the first target chart type;

[0052] Step S302, querying the chart type corresponding to the target text semantic according to the second mapping relationship to obtain the second target chart type;

[0053] Step S303, in the case that the first target chart type and the second target chart type are inconsistent, adjusting the first mapping relationship and / or the second mapping relationship so that the first target chart type and the second target chart type are consistent.

[0054] In the above embodiments, the first target chart type and the second target chart type are inconsistent. Through the triple-modal joint training framework, the system realizes bidirectional semantic mapping from text to chart type and then to data field, and significantly improves the accuracy of topic analysis. According to actual test data, the accuracy of semantic mapping from text to chart type is improved, far exceeding the traditional NLP model (the accuracy is usually less than 70%). At the same time, the framework also greatly enhances the adaptability of business scenarios. For example, in manufacturing inventory analysis, the system can automatically identify "inventory turnover rate analysis" as a time trend requirement and recommend a combination of "line chart" and "heat map" to assist in analyzing inventory fluctuations and regional distribution differences. In e-commerce sales analysis, the system can dynamically associate "user portrait" with "pie chart" and "tree chart" to display the consumer behavior characteristics of different user groups.

[0055] To improve the accuracy of topic analysis and the adaptability of business scenarios, in an optional embodiment, the step S202 further includes:

[0056] Step S2026, constructing a knowledge graph, the nodes of the knowledge graph including text semantics, the chart type and the data field, and the edges of the knowledge graph being semantic associations between the nodes;

[0057] Step S2027, analyzing the data analysis requirement text to obtain a target text semantics;

[0058] Step S2028, querying the knowledge graph according to the target text semantics to obtain the target chart type and the target data field.

[0059] In the above embodiments, in the knowledge graph mapping stage, the system further strengthens the dynamic association capability between text, chart type and data field through the dual mechanism of data dictionary and business semantic mapping and knowledge graph and semantic association. Specifically, the system first relies on the existing data dictionary of the enterprise to structurally map the meta-information of the database field (such as field name, data type, business meaning, association relationship) with the business terminology. For example, when the user inputs "inventory management", the system will automatically call the field information (such as "inventory quantity", "product category", "time dimension") in the data dictionary, and combine the business attributes of the field (such as "inventory quantity" is a numerical field, suitable for column chart or line chart display, "product category" is a text field, suitable for pie chart or tree chart display) to generate preliminary chart recommendation logic. This process not only relies on the type rule matching of the field, but also through the semantic understanding capability of the multi-modal model, the text theme (such as "inventory management") and the business semantics of the field (such as "inventory quantity" corresponds to "inventory monitoring") are deeply associated. On this basis, the system further constructs a knowledge graph to store and reason the entities such as text semantics, chart type, data field and their relationships in a graph structure. The nodes in the knowledge graph include business terminology (such as "inventory management"), data field (such as "inventory quantity"), chart type (such as "line chart"), etc., and the edges represent the semantic association between them (such as "inventory management"→"inventory quantity"→"line chart"). Through the graph neural network (GNN) or rule engine, the system can dynamically query the relationship chain in the knowledge graph to generate more accurate chart configuration schemes. For example, when the user inputs "inventory management and product distribution analysis", the system will not only match "inventory quantity" to "column chart" and "product category" to "pie chart" according to the data dictionary, but also verify the rationality of the recommended logic through the semantic path in the knowledge graph (such as "inventory management"→"product distribution"→"inventory quantity"→"column chart"), and even supplement the association of "inventory warning threshold" field and "warning heat map", so as to generate a structured chart configuration scheme. This collaborative mechanism of data dictionary and knowledge graph not only improves the system's ability to automatically identify and semantically analyze the database field, but also realizes the rapid adaptation to new fields and new business scenarios through the dynamic expansion capability of the graph structure. For example, when the enterprise adds a "inventory warning threshold" field, the system can automatically analyze its business meaning (such as "inventory safety line") through the data dictionary, and associate it to chart types such as "warning heat map" through the reasoning capability of the knowledge graph, without manual intervention to update the recommended logic. This combination of technologies enables the system to convert the static data assets of the database into dynamic semantic resources, significantly enhancing the accuracy, flexibility and adaptability of the theme analysis.The database used above has been cleaned, transformed and loaded by ETL to generate structured data for AI models. For example, fields in the database (such as inventory quantity) are converted into standardized numerical data, or business tags of fields (such as "inventory monitoring" and "trend analysis") are extracted for use by semantic models.

[0060] In order to match the template, in an optional embodiment, the above step S203 comprises:

[0061] Step S2031, match the target chart type with the layout structure of the template to obtain the target template, wherein the layout structure corresponds to the chart type of the core KPI, the chart type of the spatial distribution, and the chart type of the time trend.

[0062] In the above embodiment, the system realizes efficient conversion from the theme analysis result to the visualization scheme through a dynamic layout engine, and combines the two core modules of template matching and visual optimization to ensure that the layout scheme not only conforms to the industry best practices, but also meets the extreme needs of users for visual focus and information transmission efficiency. Specifically, the system has multiple industry template libraries built-in, covering manufacturing, retail, finance, logistics and other typical scenarios, and each template includes a standardized layout structure (such as "core KPI + spatial distribution + time trend") and chart type combination. For example, for the "warehouse monitoring" theme, the template library will recommend a layout scheme of "core KPI (total inventory, turnover rate) + spatial distribution (warehouse heat map) + time trend (inventory change line chart)". The system identifies the core entities (such as "inventory" and "warehouse") and analysis intentions (such as "monitoring" and "trend analysis") of the user input theme through a multi-modal joint training framework (text-chart-data triple modal alignment), and matches the most suitable layout structure from the template library, while supporting dynamic expansion. When the user adds a new business requirement (such as "inventory warning"), the system can automatically expand the template library to generate a combination scheme of "warning heat map + trend line chart".

[0063] In order to optimize the data chart, in an optional embodiment, the above step S204 comprises:

[0064] Step S2041, obtaining data corresponding to the target data field and filling the target template to obtain an initial data chart;

[0065] Step S2042, using AI to analyze the initial data chart to obtain an index transformation trend and a data chart adjustment strategy, wherein the index transformation trend is a transformation trend of an index corresponding to data of the data chart, and the data chart adjustment strategy is a chart type transformation strategy and a simplified data structure strategy.

[0066] Step S2043, adjusting the initial data chart according to the above data chart adjustment strategy to obtain an adjusted data chart;

[0067] Step S2044, adding the index transformation trend to the corresponding adjusted data chart to obtain the data chart.

[0068] In the above embodiment, the intelligent analysis report generates an AI analysis report structure, the chart intelligent diagnosis type compliance check (such as the time series data misuse pie chart early warning), the data density suggestion (the scatter plot data point < 50 recommends the table form), the business insight mines the key trend: the inventory turnover rate decreases by 15% (triggering the root cause analysis), the optimization opinion: delete the redundant field 1ez90s9yi2ww00, to simplify the data structure.

[0069] In order to facilitate data analysis, in an optional embodiment, the above step S205 includes:

[0070] Step S2051, the visual importance weight of the data chart is calculated by using a dynamic layout algorithm based on an attention mechanism;

[0071] Step S2052, the position of the data chart is set according to the visual importance weight, and the data analysis large screen is obtained, so that the data chart with the visual importance weight greater than a predetermined weight threshold is located in a visual focus area.

[0072] In the above implementation, in terms of visual optimization, the system adopts a dynamic layout algorithm based on attention mechanism (F-Layout) to dynamically adjust the layout structure by calculating the visual importance weight of the chart, ensuring that the core indicators occupy more than 60% of the visual focus area. Core indicator identification relies on a multi-modal model (such as a CLIP-like alignment model) to semantically analyze key indicators (such as "inventory turnover rate" and "inbound quantity") in user input text and assign them high weights. The F-Layout algorithm combines constraint solvers and attention mechanisms to adjust chart positions and sizes in real time, such as placing the "inventory turnover rate" chart in the top left corner of the screen (visual focus area) and enlarging its size, while dividing the "core KPI" "spatial distribution" "time trend" into layers through grid division (such as three-column layout) to avoid information overload. This algorithm supports multi-terminal adaptation (PC / large screen / mobile terminal), ensuring that core indicators always occupy the visual focus on different screen sizes. Through dynamic matching driven by an industry template library and the F-Layout algorithm based on attention mechanisms, the system realizes intelligent optimization of the entire process from theme analysis to visualization solutions. This technical breakthrough breaks through the limitations of traditional layout methods, not only improving the flexibility and accuracy of layout solutions, but also significantly enhancing information transmission efficiency. For example, in manufacturing inventory analysis, the system can automatically recommend a combination of "line chart (time trend) + heat map (regional distribution)", and place the "inventory turnover rate" chart in the visual focus area through the F-Layout algorithm to assist in rapid decision-making; in e-commerce sales analysis, dynamically match the "pie chart (user group distribution) + tree chart (consumer behavior hierarchy)" layout, and optimize chart positions through attention mechanisms to improve information readability. At the same time, the system trains lightweight models through knowledge distillation techniques (such as TinyML) to ensure that the dynamic layout algorithm runs efficiently on low-power devices, further reducing deployment costs. Ultimately, the intelligent layout suggestion generation module provides users with more intelligent and flexible visualization solutions through multi-modal joint training and dynamic optimization algorithms, significantly improving the user experience and business value of the data analysis platform. In summary, different themes can match different templates, and the main content can be focused on the core area through the corresponding algorithm, increasing the user optimization experience. The final large screen generated by intelligence not only can provide real-time feedback in the system, but also can realize real-time updating and interaction of data through dynamic binding of database data. Users can flexibly edit and modify the generated large screen, such as adjusting chart styles, adding or deleting data fields, optimizing layout and typesetting, or even customizing interaction logic through drag-and-drop components or configuration parameters. The system supports multi-level permission management to ensure controllability of operations by different role users (such as data analysts, business personnel, and administrators) at different stages. After completing the editing, users can publish the large screen to designated platforms (such as enterprise intranet, mobile terminal, or external display screen) and support multi-terminal adaptive display.The large screen after publishing can be in real-time linkage with the database, supports mechanisms such as timing refreshing and event triggering updating, and ensures that data is always synchronized with business scenarios. In addition, the system also provides version management functions, users can trace back to historical versions, compare modification records, and generate large screen usage reports through visual tools, further improving the efficiency of data analysis and the accuracy of business decisions.

[0073] The embodiment of the application further provides a data analysis large screen generation device. It should be noted that the data analysis large screen generation device of the embodiment of the application can be used to execute the data analysis large screen generation method provided by the embodiment of the application. The device is used to realize the above-mentioned embodiments and preferred embodiments, and details are not repeated. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware or a combination of software and hardware is also possible and is contemplated.

[0074] The data analysis large screen generation device provided by the embodiment of the application is introduced below.

[0075] Figure 3 is a schematic diagram of the data analysis large screen generation device according to the embodiment of the application. As Figure 3 shown, the device includes:

[0076] The acquisition unit 10 is configured to acquire an input data analysis requirement text, wherein the data analysis requirement text is a text describing a data analysis requirement.

[0077] Specifically, the user inputs the target analysis theme, i.e., the data analysis requirement text, through the interactive pop-up window, such as "inventory management and product distribution analysis".

[0078] The first matching unit 20 is configured to match the data analysis requirement text, the chart type and the data field to obtain a target chart type and a target data field, wherein the target chart type is a chart type that meets the data analysis requirement, and the target data field is the name of the data required to meet the data analysis requirement.

[0079] Specifically, the text semantics, chart type features and data field attributes input by the user are deeply associated through multi-modal joint training and contrast learning technology to form a "text→chart type→data field" bidirectional semantic mapping, thereby realizing more accurate chart recommendation and data association.

[0080] The second matching unit 30 is configured to match a chart template according to the target chart type to obtain a target template.

[0081] Specifically, the most suitable layout structure is matched from the template library, and the chart type corresponding to the layout structure of the target template includes the target chart type.

[0082] The filling unit 40 is configured to obtain data corresponding to the target data field and fill the target template to obtain a data chart.

[0083] Specifically, the data corresponding to the target data field is called to fill the target template, and the data chart is obtained.

[0084] The generating unit 50 is configured to generate a data analysis large screen according to the data chart.

[0085] Specifically, the data chart is laid out and displayed on the large screen to obtain the data analysis large screen.

[0086] In the data analysis large screen generation device, the chart type and the data field are matched according to the data analysis requirement text input by a user to obtain a chart type meeting the data analysis requirement and required data, so that a data chart is generated, the data chart is laid out and displayed on a large screen to obtain a data analysis large screen. Without developing a corresponding data analysis large screen generation script for different data analysis requirements, the workload of generating charts is greatly reduced, the data analysis efficiency is improved, and the problem of low data analysis efficiency in the prior art is solved.

[0087] To improve the accuracy of topic analysis and the adaptability to business scenarios, in an optional implementation, the first matching unit includes:

[0088] The first association module is configured to associate the text semantics with the data field through a contrast learning technology to obtain a first mapping relationship.

[0089] The second association module is configured to associate the text semantics with the target chart type through the contrast learning technology to obtain a second mapping relationship.

[0090] The first analysis module is configured to analyze the data analysis requirement text to obtain target text semantics.

[0091] The first query module is configured to query the data field corresponding to the target text semantics according to the first mapping relationship to obtain the target data field.

[0092] The determination module is configured to determine the chart type corresponding to the field type of the target data field as the target chart type, or query the chart type corresponding to the target text semantics according to the second mapping relationship to obtain the target chart type. The field type includes a text type field and a numerical type field.

[0093] In the above embodiments, in the large-screen theme intelligent analysis link, the system realizes the cross-modal semantic alignment of user input text and chart type, data field by constructing a text-chart-data triple-modal joint training framework, breaking through the limitations of traditional NLP model one-way text understanding, and significantly improving the accuracy of theme analysis and the adaptability of business scenarios. The core of the framework is to associate the user input text semantics, chart type characteristics and data field attributes through multi-modal joint training and contrastive learning technology, forming a "text→chart type→data field" bidirectional semantic mapping, so as to realize more accurate chart recommendation and data association. Specifically, the system first introduces a CLIP-like multi-modal alignment model, which aligns the user input text (such as "inventory management") and chart type (such as column chart, heat map) through contrastive learning technology. For example, when the user inputs "inventory management", the model analyzes the core entities in the text (such as "inventory" "product distribution") and analyzes the intent (such as "monitoring" "prediction"), and through the contrastive learning technology, the text semantics and chart type are semantically associated. For example, "inventory management" may be associated with "line chart" (for displaying time trend) and "heat map" (for displaying regional distribution) at the same time, rather than a single chart type. This cross-modal alignment not only relies on the direct mapping of text and chart, but also ensures the semantic consistency of text and chart through similarity calculation in the multi-modal embedding space. In the semantic deconstruction stage, the system further optimizes the semantic association of text and chart type through contrastive learning. The core idea of contrastive learning is to maximize the similarity of positive sample pairs and minimize the similarity of negative sample pairs. For example, in the positive sample pair of "inventory management" and "line chart", the model learns the semantic similarity of text and chart; while in the negative sample pair of "inventory management" and "pie chart", the model reduces the similarity of text and chart. Through this training, the system can dynamically identify the implicit chart type demand in the text and generate more suitable recommendation results for business scenarios. For example, when the user inputs "inventory turnover rate analysis", the system will combine "turnover rate" (time trend) and "analysis" (monitoring demand) in the text to automatically recommend "line chart" (to display time series changes) and "heat map" (to display regional distribution differences), rather than a single chart type, so as to meet the needs of complex business scenarios. In addition, the system further combines the output of the multi-modal model in the knowledge graph mapping stage to dynamically associate the fields in the enterprise data asset directory (such as "inventory quantity" "product category"). For example, when the system identifies "inventory quantity" as a numerical field, it will automatically recommend "column chart" or "line chart" as the adaptive chart type; while identifying "product category" as a text field, it will recommend "pie chart" or "tree chart" as the adaptive chart type.This dynamic association mechanism not only relies on the rule matching of field types and chart types, but also through the semantic understanding ability of the multi-modal model, the data fields are semantically associated with the text topics at a deeper level. For example, when the user inputs "inventory management and product distribution analysis", the system will combine the "inventory management" and "product distribution" two text topics, dynamically map the "inventory quantity" field to the "column chart", and associate the "product category" field to the "pie chart", thereby generating a structured chart configuration scheme.

[0094] In addition, the technology also has dynamic expansion capability. When the enterprise adds new data fields (such as "inventory warning threshold"), the system can automatically expand the chart type recommendation range (such as adding "warning heat map") through the semantic understanding ability of the multi-modal model, without manual intervention. This flexibility enables the system to quickly adapt to new business needs, further improving the intelligent level of data analysis.

[0095] In summary, the large-screen theme intelligent analysis link realizes the deep semantic association of text, chart and data through the triple modal joint training framework and contrast learning technology, not only breaking through the limitations of traditional NLP models, but also significantly improving the accuracy of theme analysis and the adaptability of business scenarios, providing a more intelligent and flexible user interaction experience for data analysis platforms.

[0096] In order to further improve the accuracy of theme analysis, in an optional implementation, the above apparatus further comprises:

[0097] The determination unit is configured to determine the chart type corresponding to the field type of the target data field as the target chart type, or, after querying the chart type corresponding to the target text semantic according to the second mapping relationship to obtain the target chart type, determine the chart type corresponding to the field type of the target data field as a first target chart type;

[0098] The query unit is configured to query the chart type corresponding to the target text semantic according to the second mapping relationship to obtain a second target chart type;

[0099] The adjustment unit is configured to, in the case that the first target chart type and the second target chart type are inconsistent, adjust the first mapping relationship and / or the second mapping relationship, so that the first target chart type and the second target chart type are consistent.

[0100] In the above embodiments, the first target chart type and the second target chart type are inconsistent. Through the triple-modal joint training framework, the system realizes bidirectional semantic mapping from text to chart type and then to data field, and significantly improves the accuracy of topic analysis. According to actual test data, the accuracy of semantic mapping from text to chart type is improved, far exceeding the traditional NLP model (the accuracy is usually less than 70%). At the same time, the framework also greatly enhances the adaptability of business scenarios. For example, in manufacturing inventory analysis, the system can automatically identify "inventory turnover rate analysis" as a time trend requirement and recommend a combination of "line chart" and "heat map" to assist in analyzing inventory fluctuations and regional distribution differences. In e-commerce sales analysis, the system can dynamically associate "user portrait" with "pie chart" and "tree chart" to display the consumer behavior characteristics of different user groups.

[0101] To improve the accuracy of topic analysis and the adaptability of business scenarios, in an optional embodiment, the first matching unit further includes:

[0102] A construction module is configured to construct a knowledge graph. The nodes of the knowledge graph include text semantics, chart types, and data fields. The edges of the knowledge graph are semantic associations between the nodes.

[0103] A second analysis module is configured to analyze the data analysis requirement text to obtain a target text semantics.

[0104] A second query module is configured to query the knowledge graph according to the target text semantics to obtain the target chart type and the target data field.

[0105] In the above embodiments, in the knowledge graph mapping stage, the system further strengthens the dynamic association capability between text, chart type and data field through the dual mechanism of data dictionary and business semantic mapping and knowledge graph and semantic association. Specifically, the system first relies on the existing data dictionary of the enterprise to structurally map the meta-information of the database field (such as field name, data type, business meaning, association relationship) with the business terminology. For example, when the user inputs "inventory management", the system will automatically call the field information (such as "inventory quantity", "product category", "time dimension") in the data dictionary, and combine the business attributes of the field (such as "inventory quantity" is a numerical field, suitable for column chart or line chart display, "product category" is a text field, suitable for pie chart or tree chart display) to generate preliminary chart recommendation logic. This process not only relies on the type rule matching of the field, but also through the semantic understanding capability of the multi-modal model, the text theme (such as "inventory management") and the business semantics of the field (such as "inventory quantity" corresponds to "inventory monitoring") are deeply associated. On this basis, the system further constructs a knowledge graph to store and reason the entities such as text semantics, chart type, data field and their relationships in a graph structure. The nodes in the knowledge graph include business terminology (such as "inventory management"), data field (such as "inventory quantity"), chart type (such as "line chart"), etc., and the edges represent the semantic association between them (such as "inventory management"→"inventory quantity"→"line chart"). Through the graph neural network (GNN) or rule engine, the system can dynamically query the relationship chain in the knowledge graph to generate more accurate chart configuration schemes. For example, when the user inputs "inventory management and product distribution analysis", the system will not only match "inventory quantity" to "column chart" and "product category" to "pie chart" according to the data dictionary, but also verify the rationality of the recommended logic through the semantic path in the knowledge graph (such as "inventory management"→"product distribution"→"inventory quantity"→"column chart"), and even supplement the association of "inventory warning threshold" field and "warning heat map", so as to generate a structured chart configuration scheme. This collaborative mechanism of data dictionary and knowledge graph not only improves the system's ability to automatically identify and semantically analyze the database field, but also realizes the rapid adaptation to new fields and new business scenarios through the dynamic expansion capability of the graph structure. For example, when the enterprise adds a "inventory warning threshold" field, the system can automatically analyze its business meaning (such as "inventory safety line") through the data dictionary, and associate it to chart types such as "warning heat map" through the reasoning capability of the knowledge graph, without manual intervention to update the recommended logic. This combination of technologies enables the system to convert the static data assets of the database into dynamic semantic resources, significantly enhancing the accuracy, flexibility and adaptability of the theme analysis.The database used above has been cleaned, converted and loaded by ETL to generate structured data for AI models. For example, fields in the database (such as inventory quantity) are converted into standardized numerical data, or business tags of fields (such as "inventory monitoring" and "trend analysis") are extracted for use by semantic models.

[0106] In order to match the template, in an optional embodiment, the second matching unit comprises:

[0107] A matching module is configured to match the target chart type with a layout structure of a template to obtain the target template, wherein the layout structure corresponds to the chart type, which includes the chart type corresponding to the core KPI, the chart type corresponding to the spatial distribution, and the chart type corresponding to the time trend.

[0108] In the above embodiment, the system realizes efficient conversion from the theme analysis result to the visualization scheme through the dynamic layout engine, and combines the two core modules of template matching and visual optimization to ensure that the layout scheme not only conforms to the industry best practices, but also meets the extreme needs of users for visual focus and information transmission efficiency. Specifically, the system has multiple industry template libraries built-in, covering manufacturing, retail, finance, logistics and other typical scenarios in multiple fields. Each template includes a standardized layout structure (such as "core KPI + spatial distribution + time trend") and chart type combination. For example, for the "warehouse monitoring" theme, the template library will recommend a layout scheme of "core KPI (total inventory, turnover rate) + spatial distribution (warehouse heat map) + time trend (inventory change line chart)". The system identifies the core entities (such as "inventory" and "warehouse") and analysis intentions (such as "monitoring" and "trend analysis") of the user input theme through a multi-modal joint training framework (text-chart-data triple modal alignment), and matches the most suitable layout structure from the template library, while supporting dynamic expansion. When the user adds a new business requirement (such as "inventory warning"), the system can automatically expand the template library to generate a combination scheme of "warning heat map + trend line chart".

[0109] In order to optimize the data chart, in an optional embodiment, the filling unit comprises:

[0110] An obtaining module is configured to obtain data corresponding to the target data field and fill the target template to obtain an initial data chart.

[0111] An analysis module is configured to analyze the initial data chart using AI to obtain an index transformation trend and a data chart adjustment strategy, wherein the index transformation trend is a transformation trend of an index corresponding to data of the data chart, and the data chart adjustment strategy is a chart type transformation strategy and a simplified data structure strategy.

[0112] An adjusting module is configured to adjust the initial data chart according to the data chart to obtain an adjusted data chart;

[0113] An adding module is configured to add the index transformation trend to the corresponding adjusted data chart to obtain the data chart.

[0114] In the above embodiment, the intelligent analysis report generates an AI analysis report structure, the chart intelligent diagnosis type compliance check (such as the time series data misuse pie chart early warning), the data density suggestion (the scatter plot data point < 50 recommended table form), the business insight mines the key trend: the inventory turnover rate decreases by 15% (triggering the root cause analysis), and the optimization suggestion: delete the redundant field 1ez90s9yi2ww00 to simplify the data structure.

[0115] In order to facilitate data analysis, in an optional embodiment, the generating unit comprises:

[0116] A calculating module is configured to calculate the visual importance weight of the data chart by using a dynamic layout algorithm based on an attention mechanism;

[0117] A setting module is configured to set the position of the data chart according to the visual importance weight to obtain the data analysis large screen, so that the data chart with the visual importance weight greater than a predetermined weight threshold is located in a visual focus area.

[0118] In the above implementation, in terms of visual optimization, the system adopts a dynamic layout algorithm based on attention mechanism (F-Layout) to dynamically adjust the layout structure by calculating the visual importance weight of the chart, ensuring that the core indicators occupy more than 60% of the visual focus area. Core indicator identification relies on a multi-modal model (such as a CLIP-like alignment model) to semantically analyze key indicators (such as "inventory turnover rate" and "inbound quantity") in user input text and assign them high weights. The F-Layout algorithm combines constraint solvers and attention mechanisms to adjust chart positions and sizes in real time, such as placing the "inventory turnover rate" chart in the top left corner of the screen (visual focus area) and enlarging its size, while dividing the "core KPI" "spatial distribution" "time trend" into layers through grid division (such as three-column layout) to avoid information overload. This algorithm supports multi-terminal adaptation (PC / large screen / mobile terminal), ensuring that core indicators always occupy the visual focus on different screen sizes. Through dynamic matching driven by an industry template library and the F-Layout algorithm based on attention mechanisms, the system realizes intelligent optimization of the entire process from theme analysis to visualization solutions. This technical breakthrough breaks through the limitations of traditional layout methods, not only improving the flexibility and accuracy of layout solutions, but also significantly enhancing information transmission efficiency. For example, in manufacturing inventory analysis, the system can automatically recommend a combination of "line chart (time trend) + heat map (regional distribution)", and place the "inventory turnover rate" chart in the visual focus area through the F-Layout algorithm to assist in rapid decision-making; in e-commerce sales analysis, dynamically match the "pie chart (user group distribution) + tree chart (consumer behavior hierarchy)" layout, and optimize chart positions through attention mechanisms to improve information readability. At the same time, the system trains lightweight models through knowledge distillation techniques (such as TinyML) to ensure that the dynamic layout algorithm runs efficiently on low-power devices, further reducing deployment costs. Ultimately, the intelligent layout suggestion generation module provides users with more intelligent and flexible visualization solutions through multi-modal joint training and dynamic optimization algorithms, significantly improving the user experience and business value of the data analysis platform. In summary, different themes can match different templates, and the main content can be focused on the core area through the corresponding algorithm, increasing the user optimization experience. The final large screen generated by intelligence not only can provide real-time feedback in the system, but also can realize real-time updating and interaction of data through dynamic binding of database data. Users can flexibly edit and modify the generated large screen, such as adjusting chart styles, adding or deleting data fields, optimizing layout and typesetting, or even customizing interaction logic through drag-and-drop components or configuration parameters. The system supports multi-level permission management to ensure controllability of operations by different role users (such as data analysts, business personnel, and administrators) at different stages. After completing the editing, users can publish the large screen to designated platforms (such as enterprise intranet, mobile terminal, or external display screen) and support multi-terminal adaptive display.The large screen after publishing can be linked with the database in real time, supports mechanisms such as timing refreshing and event triggering updating, and ensures that data is always synchronized with business scenarios. In addition, the system also provides version management functions, users can trace back to historical versions, compare modification records, and generate large screen usage reports through visual tools, further improving the efficiency of data analysis and the accuracy of business decisions.

[0119] The generation device of the data analysis large screen includes a processor and a memory, the acquisition unit, the first matching unit, the second matching unit, the filling unit and the generation unit are all stored in the memory as program units, and the corresponding functions are realized by the processor executing the program units stored in the memory. The modules are located in the same processor; or, the modules are located in different processors in any combination.

[0120] The processor includes a core, and the core retrieves the corresponding program unit from the memory. The core can be set to one or more, and the efficiency of data analysis in the prior art can be improved by adjusting the core parameters.

[0121] The memory can include non-permanent memory in a computer readable medium, random access memory (RAM) and / or non-volatile memory such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0122] The embodiment of the application provides a computer readable storage medium, and the computer readable storage medium includes a stored program, wherein the program controls the device where the computer readable storage medium is located to execute the generation method of the data analysis large screen when the program runs.

[0123] Specifically, the generation method of the data analysis large screen includes:

[0124] Step S201, acquiring an input data analysis requirement text, the data analysis requirement text is a text describing data analysis requirements;

[0125] Specifically, the user inputs the target analysis theme, i.e. the data analysis requirement text, through an interactive pop-up window, such as "inventory management and product distribution analysis".

[0126] Step S202, matching the data analysis requirement text, the chart type and the data field to obtain a target chart type and a target data field, the target chart type is the chart type meeting the data analysis requirement, and the target data field is the name of the data required to meet the data analysis requirement;

[0127] Specifically, through multi-modal joint training and contrast learning technology, the text semantics, chart type features and data field attributes input by the user are deeply associated to form a bidirectional semantic mapping of "text→chart type→data field", so that more accurate chart recommendation and data association are realized.

[0128] Step S203, matching a chart template according to the target chart type, to obtain a target template;

[0129] Specifically, the most suitable layout structure is matched from the template library, and the chart type corresponding to the layout structure of the target template includes the target chart type.

[0130] Step S204, obtaining data corresponding to the target data field and filling the target template to obtain a data chart;

[0131] Specifically, the data corresponding to the target data field is called to fill the target template, so that the data chart is obtained.

[0132] Step S205, generating a data analysis large screen according to the data chart.

[0133] Specifically, the data chart is laid out and displayed on the large screen to obtain the data analysis large screen.

[0134] The embodiment of the application provides a processor for running a program, wherein the processor is used for running the program, and the program is used for executing the generation method of the data analysis large screen.

[0135] Specifically, the generation method of the data analysis large screen comprises:

[0136] Step S201, obtaining an input data analysis requirement text, wherein the data analysis requirement text is a text describing a data analysis requirement;

[0137] Specifically, the user inputs a target analysis theme, i.e., the data analysis requirement text, through an interactive pop-up window, such as "inventory management and product distribution analysis".

[0138] Step S202, matching the data analysis requirement text, a chart type and a data field to obtain a target chart type and a target data field, wherein the target chart type is the chart type meeting the data analysis requirement, and the target data field is the name of the data required to meet the data analysis requirement;

[0139] Specifically, through multi-modal joint training and contrast learning technology, the text semantics, chart type features and data field attributes input by the user are deeply associated to form a bidirectional semantic mapping of "text→chart type→data field", so that more accurate chart recommendation and data association are realized.

[0140] Step S203, matching a chart template according to the target chart type to obtain a target template;

[0141] Specifically, the most suitable layout structure is matched from the template library, and the chart type corresponding to the layout structure of the target template includes the target chart type.

[0142] Step S204, obtaining data corresponding to the target data field and filling the target template to obtain a data chart;

[0143] Specifically, the data corresponding to the target data field is called to fill the target template, and the data chart is obtained.

[0144] Step S205, generating a data analysis large screen according to the data chart.

[0145] Specifically, the data chart is typeset and displayed on the large screen to obtain the data analysis large screen.

[0146] The embodiment of the application provides a data analysis system, which comprises a processor, a memory, and a program stored on the memory and executable on the processor, and the processor implements at least the following steps when executing the program:

[0147] Step S201, obtaining an input data analysis requirement text, wherein the data analysis requirement text is a text describing a data analysis requirement;

[0148] Specifically, the user inputs a target analysis theme, i.e., the data analysis requirement text, through an interactive pop-up window, such as "inventory management and product distribution analysis".

[0149] Step S202, matching the data analysis requirement text, a chart type, and a data field to obtain a target chart type and a target data field, wherein the target chart type is a chart type meeting the data analysis requirement, and the target data field is the name of data required to meet the data analysis requirement;

[0150] Specifically, the text semantics input by the user, the chart type features, and the data field attributes are deeply associated through multi-modal joint training and contrast learning technology to form a bidirectional semantic mapping of "text→chart type→data field", so that more accurate chart recommendation and data association are realized.

[0151] Step S203, matching a chart template according to the target chart type to obtain a target template;

[0152] Specifically, the most suitable layout structure is matched from the template library, and the chart type corresponding to the layout structure of the target template includes the target chart type.

[0153] Step S204, obtaining the data corresponding to the target data field and filling the target template to obtain a data chart;

[0154] Specifically, the data corresponding to the target data field is called to fill the target template, and the data chart is obtained.

[0155] Step S205, generating a data analysis large screen according to the data chart.

[0156] Specifically, the data chart is laid out and displayed on the large screen to obtain the data analysis large screen.

[0157] The device herein can be a server, a PC, a PAD, a mobile phone, etc.

[0158] The application also provides a computer program product adapted to execute the program of at least the following method steps when executed on a data processing device:

[0159] Step S201, obtaining an input data analysis requirement text, the data analysis requirement text being a text describing a data analysis requirement;

[0160] Specifically, the user inputs the target analysis theme, i.e., the data analysis requirement text, through an interactive pop-up window, such as "inventory management and product distribution analysis".

[0161] Step S202, matching the data analysis requirement text, the chart type and the data field to obtain a target chart type and a target data field, the target chart type being the chart type meeting the data analysis requirement, and the target data field being the name of the data required to meet the data analysis requirement;

[0162] Specifically, the text semantics, chart type features and data field attributes input by the user are deeply associated through multi-modal joint training and contrast learning technology to form a bidirectional semantic mapping of "text→chart type→data field", thereby realizing more accurate chart recommendation and data association.

[0163] Step S203, matching a chart template according to the target chart type to obtain a target template;

[0164] Specifically, the most suitable layout structure is matched from the template library, and the layout structure of the target template corresponds to a chart type including the target chart type.

[0165] Step S204, obtaining the data corresponding to the target data field and filling the target template to obtain a data chart;

[0166] Specifically, the data corresponding to the target data field is called to fill the target template, and the data chart is obtained.

[0167] In step S205, the data analysis large screen is generated according to the data chart.

[0168] Specifically, the data chart is laid out and displayed on the large screen to obtain the data analysis large screen.

[0169] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, and can be realized by program codes executable by the computing devices, so that they can be stored in storage devices and executed by the computing devices, and in some cases, the steps shown or described can be executed in different order, or they can be made into individual integrated circuit modules, or multiple modules or steps can be made into a single integrated circuit module. Therefore, the present application is not limited to any specific combination of hardware and software.

[0170] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can be in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0171] The present application is described with reference to flowcharts and / or block diagrams according to the methods, devices (systems), and computer program products of the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to produce a machine, so that the instructions executed by the computer or other programmable data processing devices produce a device that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks. Figure 1 The devices that implement the functions specified in one or more flows and / or blocks.

[0172] These computer program instructions can also be stored in a computer-readable memory that can guide the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including instruction devices that implement the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks. Figure 1 The devices that implement the functions specified in one or more flows and / or blocks.

[0173] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 Figure 1 The flowchart blocks or blocks in the flowcharts represent a sequence of steps or a plurality of steps of the functions specified in the flowchart block or blocks and / or the block or blocks.

[0174] In one typical configuration, the computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0175] The memory can include non-persistent memory and / or volatile memory, such as random access memory (RAM) and / or cache memory, non-volatile memory, such as read-only memory (ROM), EPROM, and / or flash memory, etc. The memory is an example of computer readable media.

[0176] Computer readable media includes permanent and non-permanent, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD), or other optical storage, magnetic cassette, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer readable media does not include transitory media, such as modulated data signals and carrier waves.

[0177] The technical features of the above-described embodiments can be combined in any manner. In order to make the description concise, all possible combinations of the technical features in the above-described embodiments are not described, however, as long as the combinations of the technical features do not contradict each other, it should be considered that they are within the scope of the present specification.

[0178] ​It should also be noted that the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0179] From the above description, it can be seen that the above-mentioned embodiments of the present application achieve the following technical effects:

[0180] 1) In the data analysis large screen generation method of the present application, the data analysis requirement text input by the user is matched with the chart type and data field to obtain the chart type meeting the data analysis requirement and the required data, so as to generate a data chart, the data chart is laid out and displayed on a large screen to obtain a data analysis large screen. There is no need to develop a corresponding data analysis large screen generation script for different data analysis requirements, which greatly reduces the repeated development workload of generating charts, improves the data analysis efficiency, and solves the problem of low data analysis efficiency in the prior art.

[0181] 2) In the data analysis large screen generation device of the present application, the data analysis requirement text input by the user is matched with the chart type and data field to obtain the chart type meeting the data analysis requirement and the required data, so as to generate a data chart, the data chart is laid out and displayed on a large screen to obtain a data analysis large screen. There is no need to develop a corresponding data analysis large screen generation script for different data analysis requirements, which greatly reduces the repeated development workload of generating charts, improves the data analysis efficiency, and solves the problem of low data analysis efficiency in the prior art.

[0182] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. For those skilled in the art, the present application can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating a data analysis dashboard, characterized in that, The method comprises the following steps: obtaining an input data analysis requirement text, the data analysis requirement text being a text describing a data analysis requirement; matching the data analysis requirement text, a chart type and a data field to obtain a target chart type and a target data field, the target chart type being a chart type meeting the data analysis requirement, and the target data field being a name of data required to meet the data analysis requirement; matching a chart template according to the target chart type to obtain a target template; obtaining data corresponding to the target data field and filling the target template to obtain a data chart; generating a data analysis large screen according to the data chart.

2. The method of claim 1, wherein, The matching of the data analysis requirement text, the chart type and the data field to obtain the target chart type and the target data field comprises the following steps: associating a text semantic with the data field through a contrast learning technology to obtain a first mapping relationship; associating the text semantic with the target chart type through the contrast learning technology to obtain a second mapping relationship; parsing the data analysis requirement text to obtain a target text semantic; querying the data field corresponding to the target text semantic according to the first mapping relationship to obtain the target data field; determining the chart type corresponding to the field type of the target data field as the target chart type, or querying the chart type corresponding to the target text semantic according to the second mapping relationship to obtain the target chart type, the field type comprising a text type field and a numerical type field.

3. The method of claim 2, wherein, After the chart type corresponding to the field type of the target data field is determined as the target chart type, or the chart type corresponding to the target text semantic is queried according to the second mapping relationship to obtain the target chart type, the method further comprises the following steps: determining the chart type corresponding to the field type of the target data field as a first target chart type; querying the chart type corresponding to the target text semantic according to the second mapping relationship to obtain a second target chart type; in the case where the first target chart type and the second target chart type are inconsistent, adjusting the first mapping relationship and / or the second mapping relationship so that the first target chart type and the second target chart type are consistent.

4. The method of claim 1, wherein, The matching of the data analysis requirement text, the chart type and the data field to obtain the target chart type and the target data field comprises the following steps: constructing a knowledge graph, nodes of the knowledge graph comprising a text semantic, the chart type and the data field, and edges of the knowledge graph being semantic associations between the nodes; parsing the data analysis requirement text to obtain a target text semantic; querying the knowledge graph according to the target text semantic to obtain the target chart type and the target data field.

5. The method of claim 1, wherein, The matching of the chart template according to the target chart type to obtain the target template comprises the following steps: The target chart type is matched with a layout structure of a template, and the target template is obtained, wherein the chart type corresponding to the layout structure includes the chart type corresponding to the core KPI, the chart type corresponding to the spatial distribution, and the chart type corresponding to the time sequence trend.

6. The method of claim 1, wherein, The data corresponding to the target data field is obtained and filled into the target template to obtain a data chart, including: The data corresponding to the target data field is obtained and filled into the target template to obtain an initial data chart. An AI is used to analyze the initial data chart to obtain an index transformation trend and a data chart adjustment strategy, the index transformation trend is a transformation trend of an index corresponding to data of the data chart, and the data chart adjustment strategy is a chart type transformation strategy and a simplified data structure strategy. The initial data chart is adjusted according to the data chart adjustment strategy to obtain an adjusted data chart. The index transformation trend is added to the corresponding adjusted data chart to obtain the data chart.

7. The method according to any one of claims 1 to 6, characterized in that, A data analysis large screen is generated according to the data chart, including: A dynamic layout algorithm based on an attention mechanism is used to calculate a visual importance weight of the data chart. The position of the data chart is set according to the visual importance weight to obtain the data analysis large screen, so that the data chart with the visual importance weight greater than a predetermined weight threshold is located in a visual focus area.

8. A data analysis large screen generation device, characterized by, including: An acquisition unit is configured to acquire an input data analysis requirement text, wherein the data analysis requirement text is a text describing a data analysis requirement. A first matching unit is configured to match the data analysis requirement text, a chart type, and a data field to obtain a target chart type and a target data field, wherein the target chart type is a chart type meeting the data analysis requirement, and the target data field is a name of data required to meet the data analysis requirement. A second matching unit is configured to match a chart template according to the target chart type to obtain a target template. A filling unit is configured to obtain data corresponding to the target data field and fill the data into the target template to obtain a data chart. A generation unit is configured to generate a data analysis large screen according to the data chart.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program controls a device where the computer-readable storage medium is located to execute the method for generating the data analysis large screen according to any one of claims 1 to 7 when the program is running.

10. A data analysis system, characterized by, including: One or more processors, memories, and one or more programs, wherein the one or more programs are stored in the memories and configured to be executed by the one or more processors, and the one or more programs include a program for executing the method for generating the data analysis large screen according to any one of claims 1 to 7.

Images