A method and apparatus for creating multi-dimensional analysis charts based on the UE engine

Abstract

This invention provides a method and apparatus for creating diversified analysis charts based on a UE engine, belonging to the field of chart technology. Based on source data, this invention can obtain a business indicator weight matrix that conforms to the business scenario, and based on the source data and requirement description information, obtain a chart type weight matrix that conforms to data characteristics and user expectations. Furthermore, by comprehensively considering the business indicator weight matrix and the chart type weight matrix, it can achieve in-depth data analysis while recommending suitable chart types to users. By converting the multidimensional matrix into executable parameters for the UE engine, it can leverage the powerful computing power of cloud servers and the powerful graphics rendering capabilities of the UE engine to provide users with more intuitive and comprehensive diversified analysis charts, thereby improving display efficiency and significantly enhancing the user experience.

Description

Technical Field

[0001] This invention relates to the field of chart technology, and in particular to a method and apparatus for creating multi-dimensional analysis charts based on a UE engine. Background Technology

[0002] Charts and graphs have long been widely used as a visual and intuitive means of presenting statistical information. With increasingly complex business operations and ever-growing data volumes, data visualization and analysis have become crucial.

[0003] Current chart generation methods typically involve developers pre-creating two-dimensional chart templates, then filling these templates with user-selected and processed target data. However, this approach often fails to provide in-depth data analysis or recommend suitable chart types to users. Furthermore, the visualization quality of two-dimensional charts is poor, making it difficult for users to fully and intuitively grasp the chart content, thus negatively impacting the user experience. Summary of the Invention

[0004] This invention provides a method for creating diversified analysis charts based on a UE engine. The technical solution adopted by this invention is as follows:

[0005] In a first aspect, the present invention provides a method for creating diversified analysis charts based on a UE engine, applicable to a user terminal, the method comprising:

[0006] Obtain the source data and requirement description information input by the user;

[0007] Based on the source data, a business indicator weight matrix is ​​determined, and based on the source data and the requirement description information, a chart type weight matrix is ​​determined.

[0008] The business indicator weight matrix and the chart type weight matrix are fused to obtain a multidimensional matrix.

[0009] Based on the multidimensional matrix, the executable parameters of the UE engine are determined and uploaded to the cloud server, so that the cloud server inputs the executable parameters into the UE engine to obtain the multidimensional analysis chart output by the UE engine, and sends the multidimensional analysis chart to the user terminal.

[0010] Receive and display the diversity analysis chart.

[0011] In one embodiment of the present invention, determining the business indicator weight matrix based on the source data includes:

[0012] Determine the business scenario corresponding to the source data, and determine the benchmark indicator weight matrix corresponding to the business scenario;

[0013] Obtain historical chart operation data of users in the business scenario, and based on the historical chart operation data, correct the benchmark indicator weight matrix to obtain the business indicator weight matrix.

[0014] In one embodiment of the present invention, the step of correcting the benchmark indicator weight matrix based on the historical chart operation data to obtain the business indicator weight matrix includes:

[0015] Based on the historical chart operation data, determine the indicator preference vector for the benchmark indicator weight matrix;

[0016] The business indicator weight matrix is ​​obtained by using a Bayesian model based on the benchmark indicator weight matrix and the indicator preference vector.

[0017] In one embodiment of the present invention, determining the chart type weight matrix based on the source data and the requirement description information includes:

[0018] Data features are extracted from the source data to obtain the data structure features and numerical distribution features of the source data, and a first recommendation type weight matrix is ​​determined based on the data structure features and numerical distribution features.

[0019] Semantic recognition is performed on the demand description information to obtain keyword combinations, and a second recommendation type weight matrix is ​​determined based on the keyword combinations.

[0020] Input the first recommendation type weight matrix and the second recommendation type weight matrix into a preset joint decision model to obtain the chart type weight matrix output by the joint decision model.

[0021] In one embodiment of the present invention, the step of matrix fusion of the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix includes:

[0022] The business indicator weight matrix and the chart type weight matrix are fused by tensor product to obtain the multidimensional matrix.

[0023] In one embodiment of the present invention, determining the executable parameters of the UE engine based on the multidimensional matrix includes:

[0024] In the chart type weight matrix of the multidimensional matrix, the top n chart types with the highest weights are extracted as preferred chart types; n is a positive integer greater than or equal to 2.

[0025] For any of the preferred chart types, the associated dimension data of the preferred chart type is extracted from the multidimensional matrix, and the associated dimension data is converted into UE engine parameters; wherein, the UE engine parameters include one or more of the following: spatial layout parameters, visual style parameters, dynamic effect parameters, physical interaction parameters, camera control parameters, and hierarchical detail parameters;

[0026] Based on the UE engine parameters corresponding to each of the n preferred chart types, the executable parameters of the UE engine are obtained.

[0027] In one embodiment of the present invention, the step of inputting the executable parameters into the UE engine to obtain the diversified analysis chart output by the UE engine includes:

[0028] In the idle resource pool of the cloud server, determine the target computing power resources corresponding to the user terminal;

[0029] The UE engine runs based on the target computing resources, so that the UE engine generates the diversified analysis chart based on the executable parameters.

[0030] In one embodiment of the present invention, after receiving and displaying the diversification analysis chart, the method further includes:

[0031] Based on real-time operation information input by the user, determine the operation type and the area to be updated in the diversified analysis chart;

[0032] Based on the visualization processing strategy corresponding to the operation type, the area to be updated is updated to obtain the updated chart.

[0033] Secondly, the present invention provides an apparatus for creating diversified analysis charts based on a UE engine, which is used in a user terminal. The apparatus includes:

[0034] The information acquisition module is used to acquire source data and requirement description information input by the user;

[0035] The weight matrix determination module is used to determine the business indicator weight matrix based on the source data, and to determine the chart type weight matrix based on the source data and the requirement description information.

[0036] The matrix fusion module is used to perform matrix fusion on the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix.

[0037] The parameter determination module is used to determine the executable parameters of the UE engine based on the multidimensional matrix, and upload the executable parameters to the cloud server so that the cloud server inputs the executable parameters into the UE engine to obtain the multidimensional analysis chart output by the UE engine, and sends the multidimensional analysis chart to the user terminal.

[0038] The chart display module is used to receive and display the multivariate analysis charts.

[0039] In one embodiment of the present invention, the weight matrix determination module includes:

[0040] The benchmark matrix determination submodule is used to determine the business scenario corresponding to the source data and to determine the benchmark indicator weight matrix corresponding to the business scenario.

[0041] The weight matrix determination submodule is used to obtain historical chart operation data of users in the business scenario, and to modify the benchmark indicator weight matrix based on the historical chart operation data to obtain the business indicator weight matrix.

[0042] In one embodiment of the present invention, the weight matrix determination submodule includes:

[0043] The preference vector determination unit is used to determine the indicator preference vector for the benchmark indicator weight matrix based on the historical chart operation data.

[0044] The weight matrix determination unit is used to obtain the business indicator weight matrix based on the benchmark indicator weight matrix and the indicator preference vector using a Bayesian model.

[0045] In one embodiment of the present invention, the weight matrix determination module includes:

[0046] The first recommendation submodule is used to extract data features from the source data to obtain the data structure features and numerical distribution features of the source data, and to determine the first recommendation type weight matrix based on the data structure features and numerical distribution features.

[0047] The second recommendation submodule is used to perform semantic recognition on the demand description information, obtain keyword combinations, and determine the weight matrix of the second recommendation type based on the keyword combinations.

[0048] The joint decision-making submodule is used to input the first recommendation type weight matrix and the second recommendation type weight matrix into a preset joint decision-making model to obtain the chart type weight matrix output by the joint decision-making model.

[0049] In one embodiment of the present invention, the matrix fusion module is specifically used to perform tensor product fusion on the business indicator weight matrix and the chart type weight matrix to obtain the multidimensional matrix.

[0050] In one embodiment of the present invention, the parameter determination module includes:

[0051] The preferred chart type determination submodule is used to extract the top n chart types with the highest weights from the chart type weight matrix in the multidimensional matrix as preferred chart types; n is a positive integer greater than or equal to 2.

[0052] The parameter conversion submodule is used to extract the associated dimension data of the preferred chart type from the multidimensional matrix for any of the preferred chart types, and convert the associated dimension data into UE engine parameters; wherein, the UE engine parameters include one or more of the following: spatial layout parameters, visual style parameters, dynamic effect parameters, physical interaction parameters, camera control parameters, and hierarchical detail parameters;

[0053] The executable parameter acquisition submodule is used to obtain the executable parameters of the UE engine based on the UE engine parameters corresponding to each of the n preferred chart types.

[0054] In one embodiment of the present invention, the apparatus for generating multivariate analysis charts based on the UE engine further includes:

[0055] The region determination module is used to determine the operation type and the region to be updated in the diversified analysis chart based on real-time operation information input by the user.

[0056] The chart update module is used to update the area to be updated based on the visualization processing strategy corresponding to the operation type, so as to obtain the updated chart.

[0057] In summary, the above-mentioned method and apparatus for creating diversified analysis charts based on the UE engine have the following technical effects:

[0058] This invention provides a method for creating diversified analysis charts based on a UE engine. First, it acquires source data and requirement description information input by the user. Then, based on the source data, it determines a business indicator weight matrix and a chart type weight matrix. Subsequently, it performs matrix fusion on the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix. Based on this multidimensional matrix, it determines the executable parameters of the UE engine. These executable parameters are then uploaded to a cloud server, allowing the cloud server to input them into the UE engine, resulting in a diversified analysis chart output by the UE engine. This chart is then sent to the user terminal. Finally, the diversified analysis chart is received and displayed. This invention can obtain a business indicator weight matrix that conforms to the business scenario based on source data, and a chart type weight matrix that conforms to data characteristics and user expectations based on source data and requirement description information. By comprehensively considering the business indicator weight matrix and the chart type weight matrix, it can recommend appropriate chart types to users while achieving in-depth data analysis. By converting the multidimensional matrix into executable parameters of the UE engine, it can leverage the powerful computing power of cloud servers and the powerful graphics rendering capabilities of the UE engine to provide users with more intuitive and comprehensive diversified analysis charts, thereby improving display efficiency and significantly enhancing the user experience. Attached Figure Description

[0059] Figure 1 This is a flowchart illustrating a method for creating diversified analysis charts based on a UE engine, as provided in an embodiment of the present invention.

[0060] Figure 2 This is a schematic diagram of the functional modules of a device for creating diversified analysis charts based on the UE engine, provided in an embodiment of the present invention. Detailed Implementation

[0061] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0062] It's important to note that traditional data visualization solutions are typically limited to two-dimensional chart templates, making it difficult to meet the diverse and personalized data analysis needs of users. Unreal Engine (UE), a powerful 3D rendering and game development engine, is widely used in game development, film and television production, architectural visualization, virtual reality (VR), augmented reality (AR), and real-time rendering, among other fields. Below is a brief introduction to the UE engine:

[0063] 1. Powerful graphics rendering capabilities

[0064] The UE engine is renowned for its realistic graphics rendering effects and supports advanced graphics features such as global illumination, real-time ray tracing, dynamic shadows, high-quality texture mapping, and sophisticated material rigs.

[0065] The engine's built-in rendering technology can generate highly realistic visual effects, presenting a near-realistic visual experience in both virtual worlds in games and interior designs in architectural visualizations.

[0066] 2. Blueprint Visual Programming

[0067] The UE engine provides Blueprints, a node-based visual programming tool that allows developers to build game logic and interactive features by dragging and connecting nodes without writing complex code.

[0068] Blueprint devices significantly lower the barrier to game development, enabling designers and artists without programming experience to quickly get started, while also facilitating rapid prototyping and iteration.

[0069] 3. C++ support and high extensibility

[0070] For projects requiring high performance and complex logic, the UE engine also supports C++ programming. Developers can use C++ to write efficient code, optimize performance, and implement complex devices and functions.

[0071] UE's architecture is highly modular and supports plug-in devices. Developers can develop and integrate various custom plug-ins to extend the engine's functionality and meet the needs of specific projects.

[0072] 4. Physics engine and animation setup

[0073] The UE engine has a powerful built-in physics engine that can simulate realistic physical effects such as collisions, gravity, and fluid dynamics, providing realistic physical interactions for games and virtual scenes.

[0074] The engine also provides advanced animation tools, supporting features such as skeletal animation, animation blueprints, and animation retargeting, enabling the creation of smooth and complex character animations.

[0075] 5. Real-time collaboration and multi-person development

[0076] The UE engine supports collaborative development among multiple people. Team members can collaborate in real time through version control devices (such as Perforce) to jointly edit and update project content.

[0077] The engine also provides real-time collaboration tools, such as Live Link, which allows animators, designers, and programmers to work synchronously in a real-time environment, improving development efficiency.

[0078] Based on this, the present invention aims to provide a method for creating diversified analytical charts based on a UE engine, breaking through the limitations of traditional tools and providing users with a comprehensive, in-depth, and personalized data visualization experience. On one hand, it can obtain a business indicator weight matrix that conforms to the business scenario based on the source data, and a chart type weight matrix that conforms to the data characteristics and user expectations based on the source data and requirement description information. Furthermore, by comprehensively considering the business indicator weight matrix and the chart type weight matrix, it can recommend suitable chart types to users while achieving in-depth data analysis. On the other hand, by converting the multidimensional matrix into executable parameters of the UE engine, it can leverage the powerful computing power of cloud servers and the powerful graphics rendering capabilities of the UE engine to provide users with more intuitive and comprehensive diversified analytical charts, thereby improving display efficiency and significantly enhancing the user experience.

[0079] Reference Figure 1 The present invention provides a method for creating diversified analysis charts based on a UE engine, applicable to a user terminal. The method may include the following steps:

[0080] S101: Obtain the source data and requirement description information input by the user.

[0081] It should be noted that, considering the limited computing power of user terminals, installing the UE engine locally may not be able to effectively support its fast and stable operation. Therefore, by deploying the UE engine to a cloud server, the powerful computing resources of the cloud server can be used to provide effective computing power support for the UE engine, thereby improving the efficiency of chart creation.

[0082] Specifically, the cloud server communicates with multiple user terminals configured on multiple business nodes, with each business node corresponding one-to-one with a user terminal. A business node represents a business unit of the target enterprise; for example, it could be a management department within the target enterprise responsible for a specific business. One or more of the business nodes can conduct different or the same business activities. This allows for the creation of diverse analytical charts for business managers or employees at different business nodes, achieving effective resource utilization.

[0083] In this embodiment, while inputting source data into the user terminal, the user can also input requirement description information into the user terminal via voice and / or text. For example, the source data can be the department's financial statements for 2024, and the requirement description information can be: data analysis of sales, product categories, and regional distribution for each quarter of 2024.

[0084] S102: Based on the source data, determine the weight matrix of business indicators, and based on the source data and requirement description information, determine the weight matrix of chart types.

[0085] In this implementation, to ensure data integrity, the source data can first undergo data cleaning operations, specifically including missing value handling, outlier detection, and non-numeric field encoding. Specifically, missing value handling can employ imputation algorithms to complete missing indicators, outlier detection can use isolated forests to identify abnormal data points, and non-numeric field encoding can employ one-hot encoding.

[0086] In this embodiment, by performing business analysis on the source data, a business indicator weight matrix can be obtained. The business indicator weight matrix is ​​a real-number matrix used to characterize the relative importance weights of different business indicators; for example, it can be represented as W. b =[w1,w2,...,w n ], its ∑w i =1, w1, w2, ..., w n Each represents one of the n different business metrics.

[0087] For example, in an e-commerce scenario, the weights of business metrics might be: sales revenue (0.4), conversion rate (0.3), average order value (0.2), and return rate (0.1).

[0088] In its specific implementation, step S102, which determines the business indicator weight matrix based on the source data, may include the following sub-steps:

[0089] S102-1: Determine the business scenario corresponding to the source data, and determine the benchmark indicator weight matrix corresponding to the business scenario.

[0090] In this embodiment, by performing business analysis on the source data, the corresponding business scenario can be determined. These business scenarios include, but are not limited to, finance, healthcare, e-commerce, and industrial manufacturing.

[0091] In this implementation, after determining the business scenario, a benchmark indicator weight matrix corresponding to the business scenario will be loaded from a pre-built knowledge base. The pre-built knowledge base stores different benchmark indicator weight matrices corresponding to different business scenarios. For example, in a financial scenario, the delinquency rate, collateral ratio, and credit score in the business indicators are 0.5, 0.3, and 0.2, respectively.

[0092] S102-2: Obtain historical chart operation data of users in business scenarios, and based on the historical chart operation data, correct the benchmark indicator weight matrix to obtain the business indicator weight matrix.

[0093] In this embodiment, considering the personalized needs of users, different business indicators may have different levels of importance in different business scenarios. Therefore, the baseline indicator weight matrix will be corrected based on the user's historical chart operation data in the business scenario to obtain a more accurate business indicator weight matrix.

[0094] In practice, we can first determine the indicator preference vector for the benchmark indicator weight matrix based on historical chart operation data; then, we can obtain the business indicator weight matrix through a Bayesian model based on the benchmark indicator weight matrix and the indicator preference vector.

[0095] It should be noted that the indicator preference vector is a matrix with the same data dimension as the benchmark indicator weight matrix. For example, if the benchmark indicator weight matrix is ​​[0.5 (delinquency rate), 0.3 (collateral rate), 0.2 (credit score)], and analysis of historical chart operation data reveals that users frequently compare delinquency rate and collateral rate, then the indicator preference vector for the benchmark indicator weight matrix is ​​determined to be [0.6, 0.4, 0]. Finally, the business indicator weight matrix is ​​calculated using a Bayesian model to be [0.58, 0.39, 0.02].

[0096] It should be noted that a Bayesian model is a probabilistic model based on Bayes' theorem. It estimates and infers the probability of unknown parameters or events by combining prior knowledge and observational data.

[0097] In this implementation, by dynamically adjusting the benchmark indicator weight matrix in real time based on user business preferences in various business scenarios, industry norms and user preferences can be balanced, improving the accuracy and personalization of the business indicator weight matrix. This business indicator weight matrix can guide the priority allocation and spatial layout of visualization elements, enabling diversified analysis charts to further highlight the business indicators that users are interested in.

[0098] In the specific implementation, the step in S102, which corrects the benchmark indicator weight matrix based on historical chart operation data to obtain the business indicator weight matrix, may include the following sub-steps:

[0099] S102-3: Extract data features from the source data to obtain the data structure features and numerical distribution features of the source data, and determine the weight matrix of the first recommendation type based on the data structure features and numerical distribution features.

[0100] In this embodiment, considering that different types of source data are usually applicable to different types of icons, the data structure features and numerical distribution features of the source data are obtained by extracting data features from the source data.

[0101] It should be noted that data structure characteristics represent the data type of the source data. Data types include, but are not limited to, time series, geospatial, classification comparison, and network relationships. Different data types correspond to different applicable chart types. For example, time series are suitable for line charts and area charts, geospatial is suitable for heat maps and 3D earth models, classification comparison is suitable for bar charts and radar charts, and network relationships are suitable for Sankey diagrams and force-directed diagrams.

[0102] It's important to note that numerical distribution feature extraction involves extracting statistical features from source data to quantify the data's distribution pattern, such as skewness, kurtosis, and entropy. These features help understand the data structure, thus more accurately recommending appropriate chart types. For example, data with high skewness may be more suitable for histograms or box plots to display the distribution, while data with high entropy may indicate a uniform category distribution, suitable for pie charts or stacked bar charts. Specifically, numerical distribution features can include multimodal distribution, uniform distribution, and long-tailed distribution. Multimodal distribution is suitable for heatmaps and violin plots to show multiple density peaks and overlapping distribution areas; uniform distribution is suitable for pie charts and donut charts, emphasizing the proportion of a single dominant category; and long-tailed distribution is suitable for logarithmic bar charts to compress extreme value ranges and improve readability.

[0103] In this embodiment, by comprehensively considering the data structure characteristics and numerical distribution characteristics, a first recommendation type weight matrix suitable for the source data can be obtained from the perspective of the source data.

[0104] S102-4: Perform semantic recognition on the demand description information to obtain keyword combinations, and determine the weight matrix of the second recommendation type based on the keyword combinations.

[0105] In this implementation, by performing semantic recognition on the demand description information, keyword combinations can be extracted, and then appropriate icon types can be recommended based on these keyword combinations. For example, when the demand description information is "Please show the changes in sales revenue in each region over time," the extracted keyword combinations include: show, region, sales revenue, time, and change.

[0106] In the specific implementation, the matching score between the keyword combination and multiple chart templates can be calculated first. Then, based on the user's chart preference information, the preference score corresponding to each of the multiple chart templates can be determined. Then, based on the preference score and the matching score, the weighted score of each chart template can be determined. Finally, based on the preset number of chart templates with the highest weighted scores, the weight matrix of the second recommendation type can be obtained.

[0107] In this embodiment, by comprehensively considering the demand description information and the user's chart preference information, a second recommendation type weight matrix that meets the user's expectations can be obtained based on the user's needs.

[0108] S102-5: Input the first recommendation type weight matrix and the second recommendation type weight matrix into the preset joint decision model to obtain the chart type weight matrix output by the joint decision model.

[0109] In this embodiment, the joint decision-making model can comprehensively consider the weight information in the two weight matrices. This joint decision-making model can be a decision model based on fusion rules, where the fusion rules can include linear weighted fusion, multiplicative fusion, and maximum value fusion, etc. Alternatively, the joint decision-making model can be a more complex machine learning model, such as a neural network or support vector machine.

[0110] In this embodiment, by comprehensively considering the weight matrix of the first recommendation type and the weight matrix of the second recommendation type, the chart type weight matrix can be accurately calculated by combining data features and user needs, thereby providing chart recommendations that better match data features and user personalized needs, thus improving the accuracy of recommendations.

[0111] S103: Perform matrix fusion on the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix.

[0112] In this embodiment, by fusing the business indicator weight matrix and the chart type weight matrix, the multidimensional matrix can more comprehensively reflect the characteristics of business data and users' preferences for chart types, thereby improving the depth and breadth of data analysis.

[0113] In practice, the business indicator weight matrix and the chart type weight matrix can be fused using tensor products to obtain a multidimensional matrix.

[0114] It's important to note that the tensor product (also known as the outer product or Cartesian product) is a mathematical operation that processes matrices (or tensors). It extends two matrices into a higher-dimensional space, thereby fusing their information. In this implementation, tensor product fusion can create a multidimensional matrix that simultaneously considers the weights of business metrics and chart types, providing richer information for data analysis and visualization.

[0115] S104: Based on the multidimensional matrix, determine the executable parameters of the UE engine and upload the executable parameters to the cloud server so that the cloud server can input the executable parameters into the UE engine, obtain the multidimensional analysis chart output by the UE engine, and send the multidimensional analysis chart to the user terminal.

[0116] In this embodiment, after obtaining the multidimensional matrix, the user terminal can convert the multidimensional matrix into executable parameters that the UE engine can recognize.

[0117] In a specific implementation, the step in S104 that determines the executable parameters of the UE engine based on a multi-dimensional matrix may include the following sub-steps:

[0118] S104-1: In the chart type weight matrix of the multidimensional matrix, extract the n chart types with the highest weights as the preferred chart types; n is a positive integer greater than or equal to 2.

[0119] In this implementation, the chart type weight matrix typically contains multiple chart types with different weights. Therefore, by selecting the top n chart types with the highest weights, we can meet users' personalized needs while reducing unnecessary parameter calculations and resource loading, thereby improving the rendering efficiency of the subsequent UE engine.

[0120] For example, when n is 2, if the weight of 3D bar chart is 0.6, the weight of heat map is 0.3, and the weight of Sankey diagram is 0.1 in the chart type weight matrix, then 3D bar chart and heat map are selected as preferred chart types.

[0121] S104-2: For any preferred chart type, extract the associated dimension data of the preferred chart type from the multidimensional matrix, and convert the associated dimension data into UE engine parameters.

[0122] It should be noted that UE engine parameters include one or more of the following: spatial layout parameters, visual style parameters, dynamic effect parameters, physical interaction parameters, camera control parameters, and hierarchical detail parameters.

[0123] In this implementation, for each preferred chart type, relevant dimension data related to the preferred chart type will be extracted from the multidimensional matrix, and then this relevant dimension data will be converted into corresponding UE engine parameters. Based on the converted UE engine parameters, the visualization effects of the chart can be created or adjusted in the UE engine, which can specifically include setting the chart size, color, animation effects, and interactive functions.

[0124] For example, in a multi-dimensional analysis scenario of medical images, when volume rendering is a preferred chart type, the chart type weights corresponding to volume rendering can be extracted from the chart type weight matrix, and the associated business data related to volume rendering can be extracted from the business indicator weight matrix. For example, the relevant associated business data can be extracted based on the correlation between organs and tissues. Then, based on the chart type weights corresponding to volume rendering and the associated business data, the associated dimension data of the preferred chart type can be obtained. Finally, the associated dimension data is converted into UE engine parameters to guide the UE engine to perform the corresponding rendering work. For example, high-weight areas are rendered with high-density red, and the outline lighting effect of related tissue boundaries is automatically enhanced.

[0125] S104-3: Based on the UE engine parameters corresponding to each of the n preferred chart types, obtain the executable parameters of the UE engine.

[0126] In this embodiment, by integrating the UE engine parameters corresponding to each of the n preferred chart types, the executable parameters of the UE engine can be obtained. By uploading the executable parameters of the UE engine to the cloud server, the powerful computing power of the cloud server can be used to support the UE engine in completing the rendering of the preferred chart types, thereby obtaining diversified analysis charts.

[0127] In this embodiment, considering that multiple user terminals connected to the cloud server can send various corresponding executable parameters to the cloud server in a short period of time, in order to ensure that each user terminal can obtain the corresponding diversified analysis charts in a timely manner, after receiving the executable parameters sent by any user terminal, the cloud server will determine the target computing power resources corresponding to the user terminal in the idle resource library of the cloud server; and run the UE engine based on the target computing power resources, so that the UE engine can generate diversified analysis charts based on the executable parameters.

[0128] In this embodiment, the cloud server can process multiple chart generation tasks in parallel when multiple user terminals access it at the same time, thereby achieving reasonable utilization of computing resources while ensuring that each user can obtain a smooth charting experience.

[0129] S105: Receive and display multivariate analysis charts.

[0130] In this embodiment, after receiving the diversification analysis chart, the user terminal can also convert the diversification analysis chart into an interactive chart format that the user terminal can recognize and display, and then display the diversification analysis chart based on the interactive chart format.

[0131] In this embodiment, by converting the diversified analysis chart into an interactive chart, users can interact with the chart through operations such as clicking, zooming, and dragging, to further explore the data or conduct data analysis.

[0132] In this implementation, on the one hand, a business indicator weight matrix that conforms to the business scenario can be obtained based on the source data, and a chart type weight matrix that conforms to the data characteristics and user expectations can be obtained based on the source data and requirement description information. Then, by comprehensively considering the business indicator weight matrix and the chart type weight matrix, the appropriate chart type can be recommended to the user while achieving in-depth data analysis. On the other hand, by converting the multidimensional matrix into executable parameters of the UE engine, the powerful computing power of the cloud server and the powerful graphics rendering capabilities of the UE engine can be utilized to provide users with more intuitive and comprehensive diversified analysis charts, thereby improving display efficiency and significantly improving the user experience.

[0133] In a feasible implementation, after S105, the method for creating a multivariate analysis chart based on the UE engine may further include the following steps:

[0134] S106: Based on real-time operation information input by the user, determine the operation type and the area to be updated in the diversified analysis chart.

[0135] In this embodiment, to ensure the smoothness of chart display during user interaction, the user terminal will update the diversified analysis chart in real time based on the real-time operation information input by the user.

[0136] In practice, operation types include, but are not limited to, data drill-down, dimension switching, filtering conditions, chart transformation, and perspective control. For example, if a user voice inputs "Drill down to North China," the corresponding operation type is data drill-down; if a user triggers a click operation to "Switch the X-axis to the time dimension," the corresponding operation type is dimension switching; and if a user voice inputs "Show projects with a profit margin > 15%," the corresponding operation type is filtering conditions.

[0137] S107: Based on the visualization processing strategy corresponding to the operation type, update the area to be updated to obtain the updated chart.

[0138] In this embodiment, different operation types correspond to different visualization processing strategies. For example, the visualization processing strategy corresponding to data drill-down may include hierarchical map expansion and sub-chart linkage; the visualization processing strategy corresponding to dimension switching may include dynamic coordinate system remapping and animation transition; the visualization processing strategy corresponding to filtering conditions may include fading out non-compliant elements and highlighting matching items; the visualization processing strategy corresponding to chart deformation may include geometric shape interpolation transformation and physical simulation; and the visualization processing strategy corresponding to viewpoint control may include smooth camera track movement and focal length adjustment.

[0139] In this embodiment, by utilizing visualization processing strategies under different operation types, incremental updates are performed on the area to be updated, enabling diversified analysis charts to quickly respond to user operations, thereby effectively improving the user experience.

[0140] Secondly, based on the same inventive concept, and referring to... Figure 2 This invention illustrates an apparatus 200 for creating multivariate analysis charts based on a UE engine, applicable to a user terminal. The apparatus 200 includes:

[0141] The information acquisition module 201 is used to acquire the source data and requirement description information input by the user;

[0142] The weight matrix determination module 202 is used to determine the weight matrix of business indicators based on the source data, and to determine the weight matrix of chart types based on the source data and the requirement description information.

[0143] The matrix fusion module 203 is used to perform matrix fusion on the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix.

[0144] The parameter determination module 204 is used to determine the executable parameters of the UE engine based on a multidimensional matrix, and upload the executable parameters to the cloud server so that the cloud server can input the executable parameters into the UE engine, obtain the multidimensional analysis chart output by the UE engine, and send the multidimensional analysis chart to the user terminal.

[0145] The chart display module 205 is used to receive and display multivariate analysis charts.

[0146] In one embodiment of the present invention, the weight matrix determination module 202 includes:

[0147] The benchmark matrix determination submodule is used to determine the business scenario corresponding to the source data and to determine the benchmark indicator weight matrix corresponding to the business scenario.

[0148] The weight matrix determination submodule is used to obtain historical chart operation data of users in business scenarios, and to correct the benchmark indicator weight matrix based on the historical chart operation data to obtain the business indicator weight matrix.

[0149] In one embodiment of the present invention, the weight matrix determination submodule includes:

[0150] The preference vector determination unit is used to determine the indicator preference vector for the benchmark indicator weight matrix based on historical chart operation data.

[0151] The weight matrix determination unit is used to obtain the business indicator weight matrix based on the benchmark indicator weight matrix and indicator preference vector using a Bayesian model.

[0152] In one embodiment of the present invention, the weight matrix determination module 202 includes:

[0153] The first recommendation submodule is used to extract data features from the source data, obtain the data structure features and numerical distribution features of the source data, and determine the weight matrix of the first recommendation type based on the data structure features and numerical distribution features.

[0154] The second recommendation submodule is used to perform semantic recognition on the demand description information, obtain keyword combinations, and determine the weight matrix of the second recommendation type based on the keyword combinations.

[0155] The joint decision-making submodule is used to input the first recommendation type weight matrix and the second recommendation type weight matrix into a preset joint decision-making model to obtain the chart type weight matrix output by the joint decision-making model.

[0156] In one embodiment of the present invention, the matrix fusion module is specifically used to perform tensor product fusion on the business indicator weight matrix and the chart type weight matrix to obtain a multidimensional matrix.

[0157] In one embodiment of the present invention, the parameter determination module 204 includes:

[0158] The preferred chart type determination submodule is used to extract the top n chart types with the highest weights from the chart type weight matrix in the multidimensional matrix as preferred chart types; n is a positive integer greater than or equal to 2.

[0159] The parameter conversion submodule is used to extract the associated dimension data of any preferred chart type from the multidimensional matrix and convert the associated dimension data into UE engine parameters. The UE engine parameters include one or more of the following: spatial layout parameters, visual style parameters, dynamic effect parameters, physical interaction parameters, camera control parameters, and hierarchical detail parameters.

[0160] The executable parameter acquisition submodule is used to obtain the executable parameters of the UE engine based on the UE engine parameters corresponding to each of the n preferred chart types.

[0161] In one embodiment of the present invention, the apparatus 200 for generating diversified analysis charts based on the UE engine further includes:

[0162] The region determination module is used to determine the operation type and the region to be updated in the diversified analysis charts based on real-time operation information input by the user.

[0163] The chart update module is used to update the area to be updated based on the visualization processing strategy corresponding to the operation type, and obtain the updated chart.

[0164] It should be noted that the specific implementation of the apparatus 200 for generating diversified analysis charts based on the UE engine in the embodiments of the present invention refers to the specific implementation of the method for generating diversified analysis charts based on the UE engine proposed in the first aspect of the present invention, and will not be repeated here.

[0165] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the flow or function according to the embodiments of the present invention is generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, a computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0166] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0167] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0168] Those skilled in the art will recognize that the sub-modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0169] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and sub-modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0170] In the several embodiments provided by this invention, it should be understood that the disclosed apparatus, devices, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of submodules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple submodules or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or submodules may be electrical, mechanical, or other forms.

[0171] The sub-modules described as separate components may or may not be physically separate. Similarly, the components shown as sub-modules may or may not be physical sub-modules; they may be located in one place or distributed across multiple network sub-modules. Some or all of the sub-modules can be selected to achieve the purpose of this embodiment's solution, depending on actual needs.

[0172] In addition, the functional sub-modules in the various embodiments of the present invention can be integrated into one processing sub-module, or each sub-module can exist physically separately, or two or more sub-modules can be integrated into one sub-module.

[0173] If a function is implemented as a software functional submodule and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0174] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for making a diversified analysis chart based on a UE engine, characterized by, The method applied to a user terminal comprises: obtaining source data and demand description information input by a user; determining a business index weight matrix based on the source data, and determining a chart type weight matrix based on the source data and the demand description information; performing matrix fusion on the business index weight matrix and the chart type weight matrix to obtain a multi-dimensional matrix; determining an executable parameter of a UE engine based on the multi-dimensional matrix, and uploading the executable parameter to a cloud server, so that the cloud server inputs the executable parameter into the UE engine to obtain a diversified analysis chart output by the UE engine, and sends the diversified analysis chart to the user terminal; receiving and displaying the diversified analysis chart; the determination of the business index weight matrix based on the source data comprises: determining a business scenario corresponding to the source data, and determining a benchmark index weight matrix corresponding to the business scenario; obtaining historical chart operation data of the user in the business scenario, and correcting the benchmark index weight matrix based on the historical chart operation data to obtain the business index weight matrix; the determination of the chart type weight matrix based on the source data and the demand description information comprises: performing data feature extraction on the source data to obtain data structure features and numerical distribution features of the source data, and determining a first recommended type weight matrix based on the data structure features and the numerical distribution features; performing semantic recognition on the demand description information to obtain a keyword combination, and determining a second recommended type weight matrix based on the keyword combination; inputting the first recommended type weight matrix and the second recommended type weight matrix into a preset joint decision model to obtain a chart type weight matrix output by the joint decision model.

2. The method of claim 1, wherein the UE engine is configured to generate a diversified analysis chart. the correction of the benchmark index weight matrix based on the historical chart operation data to obtain the business index weight matrix comprises: determining an index preference vector for the benchmark index weight matrix based on the historical chart operation data; obtaining the business index weight matrix based on the benchmark index weight matrix and the index preference vector through a Bayesian model.

3. The method of claim 1, wherein the UE engine is configured to generate a diversified analysis chart. the matrix fusion of the business index weight matrix and the chart type weight matrix to obtain the multi-dimensional matrix comprises: performing tensor product fusion on the business index weight matrix and the chart type weight matrix to obtain the multi-dimensional matrix.

4. The method of claim 1, wherein the method further comprises: the determination of the executable parameter of the UE engine based on the multi-dimensional matrix comprises: extracting the top n chart types in the chart type weight matrix in the multi-dimensional matrix as preferred chart types; n is a positive integer greater than or equal to 2; for any preferred chart type, extracting associated dimension data of the preferred chart type in the multi-dimensional matrix, and converting the associated dimension data into a UE engine parameter; wherein the UE engine parameter comprises one or more of a spatial layout parameter, a visual style parameter, a dynamic effect parameter, a physical interaction parameter, a camera control parameter and a hierarchical detail parameter. Based on n said preferred chart type corresponding to the respective UE engine parameters, the executable parameters of the UE engine are obtained.

5. The method of claim 1, wherein the method further comprises: The executable parameters are input into the UE engine, and the diversified analysis chart output by the UE engine is obtained. In the idle resource library of the cloud server, the target computing resource corresponding to the user terminal is determined. Based on the target computing resource, the UE engine is run to enable the UE engine to generate the diversified analysis chart based on the executable parameters.

6. The method of claim 1, wherein the method further comprises: After receiving and displaying the diversified analysis chart, the method further comprises: Based on the real-time operation information input by the user, the operation type and the area to be updated of the diversified analysis chart are determined; Based on the visual processing strategy corresponding to the operation type, the area to be updated is updated to obtain an updated chart.

7. An apparatus for making a diversified analysis chart based on a UE engine, comprising: The device is applied to a user terminal and comprises: An information acquisition module is configured to acquire source data and demand description information input by a user; A weight matrix determination module is configured to determine a business index weight matrix based on the source data and determine a chart type weight matrix based on the source data and the demand description information; A matrix fusion module is configured to fuse the business index weight matrix and the chart type weight matrix to obtain a multidimensional matrix; A parameter determination module is configured to determine executable parameters of a UE engine based on the multidimensional matrix and upload the executable parameters to a cloud server, so that the cloud server inputs the executable parameters into the UE engine to obtain a diversified analysis chart output by the UE engine and sends the diversified analysis chart to the user terminal; A chart display module is configured to receive and display the diversified analysis chart. The weight matrix determination module comprises: A reference matrix determination submodule is configured to determine a business scenario corresponding to the source data and determine a reference index weight matrix corresponding to the business scenario; A weight matrix determination submodule is configured to acquire historical chart operation data of the user in the business scenario and correct the reference index weight matrix based on the historical chart operation data to obtain the business index weight matrix; The weight matrix determination module comprises: A first recommendation submodule is configured to extract data features of the source data to obtain data structure features and numerical distribution features of the source data and determine a first recommended type weight matrix based on the data structure features and the numerical distribution features; A second recommendation submodule is configured to perform semantic recognition on the demand description information to obtain a keyword combination and determine a second recommended type weight matrix based on the keyword combination; A joint decision submodule is configured to input the first recommended type weight matrix and the second recommended type weight matrix into a preset joint decision model to obtain a chart type weight matrix output by the joint decision model.

Images