Data visualization method and system

Abstract

The invention relates to the technical field of data processing, in particular to a data visualization method and system.The system comprises a data preprocessing module, a hierarchical structure construction module, a force-oriented layout optimization module and a visualization rendering module, and the data preprocessing module, the hierarchical structure construction module, the force-oriented layout optimization module and the visualization rendering module are connected in sequence; the data preprocessing module is used for collecting and preprocessing the data; the hierarchical structure construction module is used for performing hierarchical structure construction on the preprocessed data to obtain a hierarchical tree; the force-oriented layout optimization module is used for carrying out layout optimization based on the hierarchical tree to obtain an optimized hierarchical structure; therefore, through the construction of the hierarchical structure and the layout optimization of the force-oriented algorithm, the display effect of the visual model is greatly improved, so that the relationship construction among complex data is clear, the visual effect is improved, and the user can watch more conveniently.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a data visualization method and system. Background Art

[0002] Currently, in data processing, in order to better present data to users, data is usually visualized. Data visualization is a technical means of converting a large amount of complex data into easy-to-understand graphics, images or animations. In the process of data processing and analysis, data visualization plays a crucial role. It uses visual elements, such as charts, curves, bar charts, pie charts, scatter plots and various custom graphics, etc., to display the characteristics and trends of data, so that the key information and patterns in the data can be highlighted.

[0003] However, in the aforementioned prior art, when the current data visualization means presents complex data relationships, it usually only displays through simple charts or animations. When facing huge and complex data, the visualization effect is not good, it is difficult to clearly convey the relationships between multiple data, and it is impossible to construct a clear and intuitive display of complex data relationships. Summary of the Invention

[0004] The purpose of the present invention is to provide a data visualization method and system, which solves the problem that in the prior art, when the current data visualization means presents complex data relationships, it usually only displays through simple charts or animations. When facing huge and complex data, the visualization effect is not good, it is difficult to clearly convey the relationships between multiple data, and it is impossible to construct a clear and intuitive display of complex data relationships.

[0005] To achieve the above purpose, the present invention provides a data visualization system, including a data preprocessing module, a hierarchical structure construction module, a force-directed layout optimization module and a visualization rendering module. The data preprocessing module, the hierarchical structure construction module, the force-directed layout optimization module and the visualization rendering module are connected in sequence;

[0006] The data preprocessing module is used to collect and preprocess data;

[0007] The hierarchical structure construction module is used to construct a hierarchical structure for the preprocessed data to obtain a hierarchical tree;

[0008] The force-directed layout optimization module is used to optimize the layout based on the hierarchical tree to obtain an optimized hierarchical structure;

[0009] The visualization rendering module is used to draw graphics for the optimized hierarchical structure to obtain a visualization graphic.

[0010] Among them, the data preprocessing module includes a data collection unit, a data cleaning unit, and a data formatting unit, and the data collection unit, the data cleaning unit, and the data formatting unit are connected in sequence;

[0011] The data collection unit is used to collect raw data;

[0012] The data cleaning unit is used to clean the data and remove duplicate, incorrect, or irrelevant data;

[0013] The data formatting unit is used to convert the data into a unified format.

[0014] Among them, the hierarchical structure construction module includes a node clustering unit and a hierarchical relationship determination unit, and the node clustering unit and the hierarchical relationship determination unit are connected;

[0015] The node clustering unit is used to group nodes using a clustering algorithm to form a hierarchical structure;

[0016] The hierarchical relationship determination unit is used to determine the hierarchical relationship between nodes according to the hierarchical structure and generate a hierarchical tree.

[0017] Among them, the node clustering unit includes a node input subunit, a node calculation subunit, a node update subunit, and a node iteration subunit, and the node input subunit, the node calculation subunit, the node update subunit, and the node iteration subunit are connected in sequence;

[0018] The node input subunit is used to randomly select K initial center points;

[0019] The node calculation subunit is used to calculate the distance from each node to the center point and assign it to the nearest cluster;

[0020] The node update subunit is used to update the center point of each cluster;

[0021] The node iteration subunit is used to repeat the steps of the node calculation subunit and the node update subunit until the center point no longer changes or reaches the maximum number of iterations to obtain the hierarchical structure.

[0022] Among them, the force-directed layout optimization module includes an initial layout generation unit and a layout optimization unit, and the initial layout generation unit and the layout optimization unit are connected;

[0023] The initial layout generation unit is used to input the nodes and edges of the data into the force-directed algorithm to initialize the node positions;

[0024] The layout optimization unit is used to repeatedly iterate the force-directed layout until the layout is stable or the maximum number of iterations is reached, so as to obtain an optimized hierarchy.

[0025] Among them, the layout optimization unit includes a layout node calculation subunit, a layout node update subunit, a hierarchy adjustment subunit, and a layout iteration subunit. The layout node calculation subunit, the layout node update subunit, the hierarchy adjustment subunit, and the layout iteration subunit are connected in sequence;

[0026] The layout node calculation subunit is used to calculate the gravitational and repulsive forces between nodes;

[0027] The layout node update subunit is used to update the node positions according to the magnitude and direction of the forces;

[0028] The hierarchy adjustment subunit is used to combine the hierarchy information to adjust the node positions to ensure that the high-level nodes are located in the center and the low-level nodes are distributed on the periphery, so as to obtain an optimized hierarchy;

[0029] The layout iteration subunit is used to repeat the steps of the layout node calculation subunit and the layout node update subunit until the layout is stable or the maximum number of iterations is reached.

[0030] Among them, the visualization rendering module includes a graphics drawing unit, a graphics mapping unit, an attribute encoding unit, and an interaction function unit. The graphics drawing unit, the graphics mapping unit, the attribute encoding unit, and the interaction function unit are connected in sequence;

[0031] The graphics drawing unit is used to input the optimized hierarchy into a JavaScript graphics library to draw nodes and edges, so as to obtain a visualization model;

[0032] The graphics mapping unit is used to export the visualization model from JavaScript and map it to the display screen for playback and display;

[0033] The attribute encoding unit is used to set colors and sizes according to node attributes to enhance the visualization effect;

[0034] The interaction function unit is used to add functions such as zooming, dragging, and clicking to view detailed information to the visualization model, so that it has basic interaction functions and improves the user experience.

[0035] The present invention also provides a data visualization method, which adopts the above-mentioned data visualization system, and includes the following steps:

[0036] Collect the original data through the data collection unit and preprocess the collected data;

[0037] Use a clustering algorithm to group nodes and form a hierarchical structure;

[0038] According to the hierarchical structure, determine the hierarchical relationships between nodes and generate a hierarchical tree;

[0039] Based on the hierarchical tree, perform layout optimization to obtain an optimized hierarchical structure, making the relationships between complex data clear;

[0040] Draw a graph for the optimized hierarchical structure to obtain a clear and intuitive visualization graph.

[0041] A data visualization method and system of the present invention, the data preprocessing module is used to collect and preprocess data; the hierarchical structure construction module is used to construct a hierarchical structure for the preprocessed data to obtain a hierarchical tree; the force-directed layout optimization module is used to perform layout optimization based on the hierarchical tree to obtain an optimized hierarchical structure; the visualization rendering module is used to draw a graph for the optimized hierarchical structure to obtain a visualization graph;

[0042] Thus, through the construction of the hierarchical structure and the layout optimization of the force-directed algorithm, the display effect of the visualization model is greatly improved, making the relationships between complex data clear, improving the visualization effect, and being more convenient for users to view. Brief Description of the Drawings

[0043] To more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art.

[0044] Figure 1 It is the schematic diagram of the data visualization system of the present invention.

[0045] Figure 2 It is the schematic diagram of the data preprocessing module of the present invention.

[0046] Figure 3 It is the schematic diagram of the hierarchical structure construction module of the present invention.

[0047] Figure 4 It is the schematic diagram of the node clustering unit of the present invention.

[0048] Figure 5 It is the schematic diagram of the force-directed layout optimization module of the present invention.

[0049] Figure 6 It is the schematic diagram of the layout optimization unit of the present invention.

[0050] Figure 7 It is the step flow chart of the data visualization method of the present invention.

[0051] 1 - Data pre - processing module, 101 - Data collection unit, 102 - Data cleaning unit, 103 - Data formatting unit, 2 - Hierarchical structure construction module, 201 - Node clustering unit, 2011 - Node input sub - unit, 2012 - Node calculation sub - unit, 2013 - Node update sub - unit, 2014 - Node iteration sub - unit, 202 - Hierarchical relationship determination unit, 3 - Force - directed layout optimization module, 301 - Initial layout generation unit, 302 - Layout optimization unit, 3021 - Layout node calculation sub - unit, 3022 - Layout node update sub - unit, 3023 - Hierarchical structure adjustment sub - unit, 3024 - Layout iteration sub - unit, 4 - Visualization rendering module, 401 - Graphic drawing unit, 402 - Graphic mapping unit, 403 - Attribute encoding unit, 404 - Interaction function unit. Detailed implementation manners

[0052] The embodiments of the present invention will be described in detail below. The examples of the embodiments are shown in the accompanying drawings. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as a limitation to the present invention.

[0053] Please refer to Figures 1 to 6 , the present invention provides a data visualization system, specifically including:

[0054] The data pre - processing module 1 is used to collect and pre - process data;

[0055] Specifically including:

[0056] The data collection unit 101 is used to collect raw data;

[0057] Collect the data that needs to be visualized for subsequent hierarchical structure construction and force - directed layout optimization.

[0058] The data cleaning unit 102 is used to clean the data, removing duplicate, incorrect or irrelevant data;

[0059] Pre - process the data to ensure the accuracy and consistency of the data set; the cleaned data set will be used as the input for the subsequent steps.

[0060] The data formatting unit 103 is used to convert the data into a unified format.

[0061] For unified processing, formatting is required; therefore, the cleaned data is converted into a unified format (JSON or XML) for subsequent algorithm processing and visualization rendering; the formatted data set provides a standard input for the clustering algorithm, thus entering the hierarchical structure construction stage.

[0062] The hierarchical structure construction module 2 is used to construct a hierarchical structure for the preprocessed data to obtain a hierarchical tree;

[0063] Specifically, it includes:

[0064] The node clustering unit 201 is used to group nodes using a clustering algorithm to form a hierarchical structure;

[0065] Specifically, it includes:

[0066] The node input subunit 2011 is used to randomly select K initial center points;

[0067] Randomly select K points from the dataset as the initial clustering centers. In the subsequent iterative process, the algorithm calculates the distance from each data point to these center points and assigns the data points to the cluster represented by the nearest center point, and then recalculates the center point of each cluster, taking the average value of all data points in the cluster; and then repeats this process until the center point no longer changes significantly or reaches a predetermined number of iterations.

[0068] The node calculation subunit 2012 is used to calculate the distance from each node to the center point and assign it to the nearest cluster;

[0069] Once the distances from each node to all center points are calculated, the algorithm assigns each node to the cluster represented by the center point closest to it. This process is iterative because after each assignment, the algorithm recalculates the center point of each cluster (usually the average of the coordinates of all nodes in the cluster), and then uses the new center point for the next round of distance calculation and node assignment.

[0070] The node update subunit 2013 is used to update the center point of each cluster;

[0071] The common practice of updating the cluster center point is to calculate the average value (or called the centroid) of the coordinates of all nodes in each cluster. This average value serves as the new center point, representing the central position of the cluster;

[0072] By continuously updating the cluster center point, the algorithm can gradually adjust the position and shape of each cluster, making the clustering result more accurate and stable. This helps to better reflect the true distribution and internal structure of the data; at the same time, during the iterative process, as the center point is continuously updated, the algorithm will gradually approach the true clustering structure of the data, thereby reducing the number of iterations and improving the calculation efficiency.

[0073] The node iteration subunit 2014 is used to repeat the steps of the node calculation subunit 2012 and the node update subunit 2013 until the center point no longer changes or reaches the maximum number of iterations to obtain the hierarchical structure.

[0074] By continuously adjusting the node allocation and the positions of the central points, the optimal clustering result is gradually approximated. This process is similar to optimization algorithms such as gradient descent, which iteratively updates parameters to minimize the objective function; the algorithm uses distance metric methods to evaluate the distance between nodes and central points, and assigns nodes to the nearest clusters according to the distance. This process is one of the core steps of the clustering algorithm, which determines the shape and boundary of the clusters;

[0075] Through repeated iterations, the algorithm has enough time to adjust and optimize the positions and shapes of each cluster, making the clustering result more stable and accurate. This helps to reduce the fluctuations in the clustering result caused by factors such as improper selection of the initial central points or data noise; the iterative process helps the algorithm to gradually approximate the true clustering structure of the data. As the iteration progresses, the central points will gradually converge to the natural clustering centers of the data, thus revealing the internal hierarchy and grouping information of the data

[0076] The hierarchical relationship determination unit 202 is used to determine the hierarchical relationship between nodes according to the hierarchical structure and generate a hierarchical tree.

[0077] In the hierarchical structure, each node has a parent node and several child nodes, forming a tree-like structure. This structure can clearly represent the hierarchy and association relationship between data; before generating the hierarchical tree, it is necessary to first determine the hierarchical relationship between nodes. This is usually achieved through data preprocessing and analysis. For example, in a file system, the hierarchical relationship between nodes can be determined according to the path relationship between directories and files; in an organizational structure diagram, the hierarchical relationship between nodes can be determined according to the superior-subordinate relationship between employees; once the hierarchical relationship between nodes is determined, the generation of the hierarchical tree can begin. The process of generating the hierarchical tree usually involves creating nodes and edges (or links) and connecting them according to the hierarchical relationship. In the hierarchical tree, each node represents a data entity or concept, while the edges represent the hierarchical relationship between nodes; finally, the generated hierarchical tree can be visually displayed. By graphically displaying the hierarchical tree, the hierarchical relationship and structure between data can be seen more intuitively. This helps users to better understand and analyze the data; therefore, determining the hierarchical relationship between nodes and generating the hierarchical tree can effectively improve the effect of subsequent visualization.

[0078] The force-directed layout optimization module 3 is used to optimize the layout based on the hierarchical tree to obtain an optimized hierarchical structure;

[0079] Specifically, it includes:

[0080] The initial layout generation unit 301 is used to input the nodes and edges of the data into the force-directed algorithm to initialize the node positions;

[0081] In the force-directed layout algorithm, the initial positions of nodes are usually randomly distributed. This is to ensure that the algorithm can handle various possible datasets and layout requirements. Random distribution can prevent the algorithm from getting trapped in local optimal solutions and increase the diversity of the layout. The force-directed layout algorithm is based on the principle of physical simulation, regarding nodes as particles with mass and edges as springs or elastic forces connecting the particles. After initializing the node positions, the algorithm will simulate the movement process of these particles under the action of repulsive and attractive forces. The repulsive force keeps adjacent nodes at a certain distance to avoid overlap, while the attractive force keeps the edges connecting the nodes at a certain length and direction, thus maintaining the structural integrity of the graph. The algorithm will start the iterative process. In each iteration, the algorithm calculates the repulsive and attractive forces between nodes and adjusts the positions of nodes according to the resultant force of these forces. This process will be repeated until a stable state is reached or a preset stopping condition is met (such as the number of iterations, the change in node positions is less than a threshold, etc.).

[0082] The initialized node positions serve as the initial state of the algorithm, providing a starting point for the subsequent iterative process. The algorithm will start calculating the repulsive and attractive forces between nodes based on these initial positions and gradually adjust the positions of nodes. Although the initial positions are random, they will affect the final layout result to a certain extent. Different initial positions may lead to different iterative paths and final layout forms. Therefore, the initialization process is also a source of the diversity of the algorithm results.

[0083] The layout optimization unit 302 is used to repeatedly iterate the force-directed layout until the layout is stable or the maximum number of iterations is reached, obtaining an optimized hierarchical structure.

[0084] Specifically, it includes:

[0085] The layout node calculation subunit 3012 is used to calculate the attractive and repulsive forces between nodes.

[0086] By calculating the attractive and repulsive forces between nodes, the positions of nodes can be continuously adjusted to make the entire layout reach a balanced state. This balanced state usually shows that the relative positions of nodes are reasonable, and the lengths and directions of edges meet certain aesthetic standards or actual requirements. The attractive force keeps the edges connecting the nodes at a certain length and direction, thus maintaining the structural integrity of the graph. This helps to clearly display the relationships and hierarchical structures between data during the visualization process. By continuously adjusting the attractive and repulsive forces between nodes, the layout can be made more beautiful, compact, and easy to understand. This helps users better understand and analyze data and improves the effect of data visualization.

[0087] Gravity generally represents the attractive force between nodes, which keeps the edges connecting the nodes at a certain length; the repulsive force represents the repulsive force between nodes, which keeps adjacent nodes at a certain distance to avoid overlap; in the force-directed layout algorithm, the gravity can be calculated using Hooke's law, that is, the gravity is proportional to the distance between nodes and inversely proportional to the spring constant. Specifically, the gravity can be expressed as F_ij^attract = -k * d_ij, where k represents the spring constant and d_ij represents the distance between node i and node j; the repulsive force is usually calculated using Coulomb's law, that is, the repulsive force is inversely proportional to the square of the distance between nodes. Specifically, the repulsive force can be expressed as F_ij2), where k' represents the coefficient of the repulsive force;

[0088] The resultant force on a node is the vector sum of the gravitational force and the repulsive force. By calculating the resultant force, the direction and distance of the node's movement can be determined; in each iteration, the algorithm updates the position of the node according to the magnitude and direction of the resultant force on the current node. This process will be repeated until a stable state is reached or a preset stop condition is met.

[0089] The layout node update subunit 3022 is used to update the node position according to the magnitude and direction of the force;

[0090] By continuously updating the positions of the nodes, the force-directed layout algorithm can gradually optimize the layout effect, making the relative positions of the nodes more reasonable and the lengths and directions of the edges more in line with aesthetic standards or actual requirements; the algorithm will continuously adjust the positions of the nodes until a balanced state is reached, that is, the resultant force on the node is close to zero. The layout in this state has high stability and is not easily affected by external disturbances; by updating the node positions, the force-directed layout algorithm can generate natural and beautiful graph layouts, which are convenient for users to understand and analyze data. This helps to improve the effect of data visualization and the user experience.

[0091] The hierarchical structure adjustment subunit 3023 is used to adjust the node positions in combination with the hierarchical structure information to ensure that high-level nodes are located in the center and low-level nodes are distributed on the periphery, obtaining an optimized hierarchical structure;

[0092] Before adjusting the node positions, it is necessary to analyze the hierarchical data to determine the nodes and edges at different levels. By analyzing the characteristics of the hierarchical data, the hierarchical relationship and structural characteristics of the data can be understood, providing a basis for subsequent layout adjustment; the node positions will be adjusted according to the hierarchical structure information. High-level nodes will be affected by more gravitational forces and thus be attracted to the central position; while low-level nodes will be affected by more repulsive forces and thus be distributed on the periphery of the central nodes. By continuously iterating and adjusting the node positions, an optimized layout that meets the requirements of the hierarchical structure can finally be obtained;

[0093] After finally adjusting the node positions, the layout will appear more natural and beautiful. High-level nodes are located in the center, forming a core area, while low-level nodes are distributed around the center, forming a well-defined hierarchical layout effect. This layout method conforms to people's visual habits and helps improve the user experience; achieving a clear hierarchical structure and a beautiful layout can make the data easier to read and understand. Users can quickly understand the hierarchical relationship and structural characteristics of the data by observing the layout, thereby improving the efficiency of data analysis and processing;

[0094] The layout iteration subunit 3024 is used to repeat the steps of the layout node calculation subunit 3012 and the layout node update subunit 3022 until the layout is stable or the maximum number of iterations is reached.

[0095] The process of repeated iteration helps the algorithm find a more beautiful, compact, and easy-to-understand layout method; by setting the maximum number of iterations as the stopping condition, the algorithm can complete the layout calculation within a certain time, avoiding the waste of computing resources caused by infinite iteration. At the same time, this also improves the robustness and practicality of the algorithm; thus, through repeated iteration, it helps improve the final visualization effect.

[0096] The visualization rendering module 4 is used to draw a graph of the optimized hierarchical structure to obtain a visualization graph.

[0097] Specifically, it includes:

[0098] The graph drawing unit 401 is used to input the optimized hierarchical structure into the JavaScript graph library to draw nodes and edges, obtaining a visualization model;

[0099] Input the optimized hierarchical structure data into the JavaScript graph library. The graph library will parse and process the input data, extract the information of nodes and edges; according to the parsed data, the graph library will generate corresponding graphical elements to represent nodes and edges. These graphical elements can be in shapes such as circles, rectangles, lines, etc., and the specific shapes and styles can be customized according to actual needs;

[0100] Through the JavaScript graph library, the optimized hierarchical structure can be displayed graphically, enabling users to intuitively see the hierarchical relationship and connection situation between nodes;

[0101] The graph mapping unit 402 is used to export the visualization model from JavaScript and map it to the display screen for playback and display;

[0102] After displaying the visualization model, it is convenient for users to view the data.

[0103] The attribute encoding unit 403 is used to set the color and size according to the node attributes to enhance the visualization effect;

[0104] By setting the color and size related to the attributes for the nodes, different features and categories of the nodes can be intuitively displayed; furthermore, users can quickly understand the structure and relationship of the data without carefully reading the text description, thereby improving the efficiency of information transmission.

[0105] The interaction function unit 404 is used to add functions such as zooming, dragging, and clicking to view detailed information to the visualization model, so that it has basic interaction functions and improves the user experience.

[0106] Users can add and delete custom interaction functions to the visualization model according to their own habits of observing data, thereby improving the user experience.

[0107] Please refer to Figure 7 , the present invention also provides a data visualization method, including the following steps:

[0108] S1: Collect the original data through the data collection unit 101 and preprocess the collected data;

[0109] S2: Use a clustering algorithm to group the nodes to form a hierarchical structure;

[0110] S3: Determine the hierarchical relationship between the nodes according to the hierarchical structure and generate a hierarchical tree;

[0111] S4: Perform layout optimization based on the hierarchical tree to obtain an optimized hierarchical structure, so as to build a clear relationship between complex data;

[0112] S5: Draw a graph for the optimized hierarchical structure to obtain a clearly constructed and intuitive visualization graph.

[0113] Among them, collect the original data through the data collection unit 101, preprocess the collected data; use a clustering algorithm to group the nodes to form a hierarchical structure; determine the hierarchical relationship between the nodes according to the hierarchical structure and generate a hierarchical tree; perform layout optimization based on the hierarchical tree to obtain an optimized hierarchical structure, so as to build a clear relationship between complex data; draw a graph for the optimized hierarchical structure to obtain a clearly constructed and intuitive visualization graph.

[0114] The above-disclosed are only one or more preferred embodiments of the present application, and the scope of rights of the present application cannot be limited thereby. Those of ordinary skill in the art can understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present application still fall within the scope covered by the present application.

Claims

1. A data visualization system, characterized in that: It includes a data preprocessing module, a hierarchical structure building module, a force-directed layout optimization module and a visual rendering module, wherein the data preprocessing module, the hierarchical structure building module, the force-directed layout optimization module and the visual rendering module are connected in sequence; The data preprocessing module is used to collect and preprocess the data; The hierarchical structure building module is used to build a hierarchical structure of the preprocessed data to obtain a hierarchical tree; The force-directed layout optimization module is used to perform layout optimization based on the hierarchical tree to obtain an optimized hierarchical structure; The visualization rendering module is used to draw a graph of the optimized hierarchical structure to obtain a visualization graph.

2. The data visualization system according to claim 1, characterized in that The data preprocessing module includes a data collection unit, a data cleaning unit and a data formatting unit, and the data collection unit, the data cleaning unit and the data formatting unit are connected in sequence; The data collection unit is used to collect raw data; The data cleaning unit is used to clean the data and remove duplicate, erroneous or irrelevant data; The data formatting unit is used to convert the data into a unified format.

3. The data visualization system according to claim 2, characterized in that: The hierarchical structure building module includes a node clustering unit and a hierarchical relationship determination unit, and the node clustering unit and the hierarchical relationship determination unit are connected; The node clustering unit is used to group the nodes using a clustering algorithm to form a hierarchical structure; The hierarchical relationship determination unit is used to determine the hierarchical relationship between nodes according to the hierarchical structure and generate a hierarchical tree.

4. The data visualization system according to claim 3, characterized in that: The node clustering unit includes a node input subunit, a node calculation subunit, a node update subunit and a node iteration subunit, and the node input subunit, the node calculation subunit, the node update subunit and the node iteration subunit are connected in sequence; The node input subunit is used to randomly select K initial center points; The node calculation subunit is used to calculate the distance between each node and the center point and assign it to the nearest cluster; The node updating subunit is used to update the center point of each cluster; The node iteration subunit is used to repeat the steps of the node calculation subunit and the node update subunit until the center point no longer changes or the maximum number of iterations is reached, thereby obtaining the hierarchical structure.

5. The data visualization system according to claim 4, characterized in that: The force-directed layout optimization module includes an initial layout generation unit and a layout optimization unit, and the initial layout generation unit is connected to the layout optimization unit; The initial layout generation unit is used to input the nodes and edges of the data into the force-directed algorithm to initialize the node positions; The layout optimization unit is used to repeatedly iterate the force-directed layout until the layout is stable or the maximum number of iterations is reached, thereby obtaining an optimized hierarchical structure.

6. The data visualization system according to claim 5, characterized in that: The layout optimization unit comprises a layout node calculation subunit, a layout node update subunit, a hierarchy adjustment subunit and a layout iteration subunit, wherein the layout node calculation subunit, the layout node update subunit, the hierarchy adjustment subunit and the layout iteration subunit are connected in sequence; The layout node calculation subunit is used to calculate the attraction and repulsion between nodes; The layout node updating subunit is used to update the node position according to the magnitude and direction of the force; The hierarchical structure adjustment subunit is used to adjust the node positions in combination with the hierarchical structure information to ensure that the high-level nodes are located in the center and the low-level nodes are distributed on the periphery, thereby obtaining an optimized hierarchical structure; The layout iteration subunit is used to repeat the steps of the layout node calculation subunit and the layout node update subunit until the layout is stable or the maximum number of iterations is reached.

7. The data visualization system according to claim 6, characterized in that: The visualization rendering module includes a graphics drawing unit, a graphics mapping unit, an attribute encoding unit and an interactive function unit, wherein the graphics drawing unit, the graphics mapping unit, the attribute encoding unit and the interactive function unit are connected in sequence; The graphics drawing unit is used to input the optimized hierarchical structure into a JavaScript graphics library, draw nodes and edges, and obtain a visualization model; The graphic mapping unit is used to export the visualization model from JavaScript and map it to a display screen for playback and display; The attribute encoding unit is used to set the color and size according to the node attributes to enhance the visualization effect; The interactive function unit is used to add the functions of zooming, dragging and clicking to view detailed information of the visual model, so that it has basic interactive functions and improves the user experience.

8. A data visualization method, using the data visualization system according to claim 7, characterized in that: The steps include: Collecting raw data through the data collection unit, and preprocessing the collected data; Use clustering algorithms to group nodes into hierarchical structures; According to the hierarchical structure, determine the hierarchical relationship between nodes and generate a hierarchical tree; Optimizing the layout based on the hierarchical tree to obtain an optimized hierarchical structure, so that the relationship between complex data is clearly constructed; The optimization hierarchy is graphed to obtain a clear and intuitive visualization graph.

Images