A method and system for realizing a data visualization large screen

Abstract

The present invention application provides a method and system for realizing a data visualization large screen, including: obtaining multiple data tasks and task information to be displayed; clustering each of the data tasks and constructing a tree index structure; determining the priority of each data task; obtaining the current zoom level of each sub-view area; determining the data granularity; using a progressive data loading strategy and combining the priority to load each data task in sequence; and displaying the loaded data tasks according to the task coordinates. The present invention application obtains the current zoom level of the sub-view area, thereby determining the data granularity of the data to be displayed in the current sub-view area, which can meet the differentiated needs of users for different sub-view areas and different tasks, improve the intelligence of data display on the data visualization large screen, thereby meeting the needs of flexible and complex application scenarios, and has universal applicability.

Description

Technical Field

[0001] The present invention relates to the field of visualization, and more particularly to a method and system for realizing a large-screen data visualization. Background Art

[0002] With the development of science and technology, data is showing a trend of massive amounts. In some data centers, a large amount of data is generated every day. In addition to considering how to effectively store and process data, real-time analysis and monitoring of data has also become an urgent need. At present, some data visualization screens have begun to adopt a split-screen mechanism, that is, dividing the screen view into several sub-areas. However, since these sub-areas use the same data source, they often use the same data granularity for display. In fact, users have different usage requirements for these sub-areas, and different usage requirements correspond to different granularity requirements. Therefore, if these sub-areas use the same data granularity, the data visualization screen will not be intelligent enough to display data, and it will not be able to adapt to the needs of flexible and complex application scenarios. Summary of the invention

[0003] The present invention application provides a method and system for realizing a large-screen data visualization to solve the technical problem of how to improve the intelligence level of data display and meet the needs of flexible and complex application scenarios.

[0004] In order to solve the above technical problems, the present invention provides a method for implementing a data visualization large screen, wherein the display content of the data visualization large screen includes a preset view area, and the preset view area includes multiple sub-view areas;

[0005] The implementation method includes: obtaining multiple data tasks to be displayed on the data visualization large screen and task information of each of the data tasks; clustering each of the data tasks, and building a tree index structure according to the clustering results; wherein the task information includes task coordinates; determining the priority of each of the data tasks according to the tree index structure; obtaining the current zoom level of each of the sub-view areas; according to each of the current zoom levels, determining the data granularity of the data to be displayed in each of the sub-view areas; according to each of the data granularities, in each of the sub-view areas, adopting a progressive data loading strategy and combining the priority of each of the data tasks, loading each of the data tasks in sequence; and displaying the loaded data tasks according to each of the task coordinates.

[0006] As a preferred solution, the task information also includes data density; clustering the data tasks and constructing a tree index structure according to the clustering results includes:

[0007] Performing preliminary clustering of each of the data tasks by a hierarchical clustering algorithm; wherein, in the process of preliminary clustering, the Ward link method is used to minimize the variance of the clustering clusters;

[0008] Eliminate data tasks whose data density is less than or equal to the preset density threshold in the preliminary clustering results;

[0009] The DBSCAN algorithm is used to refine the clusters corresponding to the removed data, and the ε parameter and domain density threshold of the clusters are adjusted to obtain the optimized clustering results;

[0010] Based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

[0011] As a preferred solution, the method for updating the tree index structure includes:

[0012] Calculating the data volume, statistical indicators and task type of the data task corresponding to each node; wherein the statistical indicators are obtained by analyzing the optimized clustering results;

[0013] Monitor the tree index structure, and when at least one of the data volume, statistical indicators, and task type of the data task corresponding to the lower-level node changes, update at least one of the data volume, statistical indicators, and task type of the data task corresponding to the upper-level node through incremental calculation;

[0014] Among them, before the incremental calculation update, it also includes: when the format type of the data task corresponding to the lower-level node is inconsistent with the format type of the data task corresponding to the upper-level node, the format type of the data task corresponding to the upper-level node is converted according to the format type of the data task corresponding to the lower-level node.

[0015] As a preferred solution, determining the priority of each data task according to the tree index structure includes:

[0016] Determining a hierarchical relationship of the tree index structure according to a root node of the tree index structure;

[0017] According to the hierarchical relationship, respectively calculating the cosine similarity between each node of the tree index structure and the layer in which it is located;

[0018] According to the order of cosine similarities from large to small, the importance of each node is determined, thereby determining the priority of each data task in turn.

[0019] As a preferred solution, the implementation method further includes:

[0020] Responding to interactive operations on the data visualization screen;

[0021] Using a pre-configured data mapping table, querying according to the interaction operation to determine the interaction event type;

[0022] Determine the target interactive task according to the action area of ​​the interactive operation on the data visualization screen and the task coordinates; extract the corresponding data range and data granularity according to the interactive event type;

[0023] The range parameter of the target interactive task is updated by using the extracted data range through an asynchronous update interface, and the updated target interactive task is displayed according to the extracted data granularity.

[0024] As a preferred solution, according to each data granularity, in each sub-view area, a progressive data loading strategy is adopted and combined with the priority of each data task to sequentially load each data task, including:

[0025] Build multiple loading threads, each loading thread corresponds to a subview area;

[0026] Transmitting each of the data tasks to each loading thread in sequence according to the priority;

[0027] Obtaining a data block size threshold corresponding to each of the loading threads;

[0028] When the total size of each data task in the loading thread exceeds the data block size threshold, dividing all data tasks in the loading thread into a plurality of data sets;

[0029] A progressive data loading strategy is adopted, and in each sub-view area, each data set is loaded in sequence according to the corresponding data granularity; wherein data tasks of the same data set are loaded in parallel.

[0030] Correspondingly, the present invention application also provides a data visualization large screen implementation system, wherein the display content of the data visualization large screen includes a preset view area, and the preset view area includes multiple sub-view areas;

[0031] The implementation system includes a construction module, a priority determination module, a granularity determination module, a loading module and a display module; wherein,

[0032] The construction module is used to obtain multiple data tasks to be displayed on the data visualization screen and task information of each data task; cluster each data task and construct a tree index structure according to the clustering results; wherein the task information includes task coordinates;

[0033] The priority determination module is used to determine the priority of each of the data tasks according to the tree index structure;

[0034] The granularity determination module is used to obtain the current zoom level of each sub-view area; and determine the data granularity of the data to be displayed in each sub-view area according to each current zoom level;

[0035] The loading module is used to load each of the data tasks in sequence according to each of the data granularities and in each of the sub-view areas by adopting a progressive data loading strategy and combining the priority of each of the data tasks;

[0036] The display module is used to display the loaded data tasks according to the task coordinates.

[0037] As a preferred solution, the task information also includes data density; the construction module clusters each of the data tasks and constructs a tree index structure according to the clustering results, including:

[0038] The construction module performs preliminary clustering of each of the data tasks through a hierarchical clustering algorithm; wherein, during the preliminary clustering process, the Ward link method is used to minimize the variance of the clusters;

[0039] Eliminate data tasks whose data density is less than or equal to the preset density threshold in the preliminary clustering results;

[0040] The DBSCAN algorithm is used to refine the clusters corresponding to the removed data, and the ε parameter and domain density threshold of the clusters are adjusted to obtain the optimized clustering results;

[0041] Based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

[0042] As a preferred solution, the method for updating the tree index structure includes:

[0043] Calculating the data volume, statistical indicators and task type of the data task corresponding to each node; wherein the statistical indicators are obtained by analyzing the optimized clustering results;

[0044] Monitor the tree index structure, and when at least one of the data volume, statistical indicators, and task type of the data task corresponding to the lower-level node changes, update at least one of the data volume, statistical indicators, and task type of the data task corresponding to the upper-level node through incremental calculation;

[0045] Among them, before the incremental calculation update, it also includes: when the format type of the data task corresponding to the lower-level node is inconsistent with the format type of the data task corresponding to the upper-level node, the format type of the data task corresponding to the upper-level node is converted according to the format type of the data task corresponding to the lower-level node.

[0046] As a preferred solution, the priority determination module determines the priority of each of the data tasks according to the tree index structure, including:

[0047] The priority determination module determines the hierarchical relationship of the tree index structure according to the root node of the tree index structure;

[0048] According to the hierarchical relationship, respectively calculating the cosine similarity between each node of the tree index structure and the layer in which it is located;

[0049] According to the order of cosine similarities from large to small, the importance of each node is determined, thereby determining the priority of each data task in turn.

[0050] As a preferred solution, the implementation system further includes an update module; the update module is used to:

[0051] Responding to interactive operations on the data visualization screen;

[0052] Using a pre-configured data mapping table, querying according to the interaction operation to determine the interaction event type;

[0053] Determine the target interactive task according to the action area of ​​the interactive operation on the data visualization screen and the task coordinates; extract the corresponding data range and data granularity according to the interactive event type;

[0054] The range parameter of the target interactive task is updated by using the extracted data range through an asynchronous update interface, and the updated target interactive task is displayed according to the extracted data granularity.

[0055] As a preferred solution, the loading module adopts a progressive data loading strategy in each sub-view area according to each data granularity and combines the priority of each data task to sequentially load each data task, including:

[0056] The loading module constructs multiple loading threads, each loading thread corresponds to a subview area;

[0057] Transmitting each of the data tasks to each loading thread in sequence according to the priority;

[0058] Obtaining a data block size threshold corresponding to each of the loading threads;

[0059] When the total size of each data task in the loading thread exceeds the data block size threshold, dividing all data tasks in the loading thread into a plurality of data sets;

[0060] A progressive data loading strategy is adopted, and in each sub-view area, each data set is loaded in sequence according to the corresponding data granularity; wherein data tasks of the same data set are loaded in parallel.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention application provides a method and system for implementing a large data visualization screen, wherein the display content of the large data visualization screen includes a preset view area, and the preset view area includes multiple sub-view areas; the implementation method includes: obtaining multiple data tasks to be displayed on the data visualization screen and task information of each data task; clustering each data task, and constructing a tree index structure according to the clustering results; wherein the task information includes task coordinates; determining the priority of each data task according to the tree index structure; obtaining the current zoom level of each sub-view area; determining the data granularity of the data to be displayed in each sub-view area according to each current zoom level; according to each data granularity, in each sub-view area, adopting a progressive data loading strategy and combining the priority of each data task, loading each data task in sequence; and displaying the loaded data task according to each task coordinate. The present invention obtains the current zoom level of the sub-view area to determine the data granularity of the data to be displayed in the current sub-view area, which can meet the differentiated needs of users for different sub-view areas and different tasks, improve the intelligence level of data visualization large-screen data display, and thus meet the needs of flexible and complex application scenarios, and has universal applicability; in addition, by clustering multiple data tasks to be displayed and constructing a tree index structure to determine the priority of the data task, and loading and displaying each data task in turn according to the priority, orderly loading and display of data tasks is achieved, which can further meet the user's display needs for data tasks of different priorities and further enhance the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 : A flow chart of an embodiment of a method for realizing a data visualization large screen provided by the present invention.

[0064] Figure 2 : A structural diagram of an embodiment of a system for implementing a large-screen data visualization provided by the present invention. DETAILED DESCRIPTION

[0065] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] Embodiment 1

[0067] Please refer to Figure 1 , Figure 1 The present invention provides a method for implementing a large data visualization screen, wherein the display content of the large data visualization screen includes a preset view area, and the preset view area includes multiple sub-view areas.

[0068] In this embodiment, the method for implementing the data visualization large screen can be applied to computer devices, including but not limited to smart phones, laptops, tablet computers, desktop computers, and physical servers and cloud servers connected to display units.

[0069] The method for implementing the data visualization large screen includes steps S1 to S3, wherein the specific contents of each step are as follows:

[0070] Step S1, obtaining multiple data tasks to be displayed on the data visualization screen and task information of each data task, clustering each data task, and constructing a tree index structure according to the clustering result. The task information includes task coordinates.

[0071] In this step, the data visualization screen includes a preset view area, which may be in the form of a window. The non-preset view area of ​​the data visualization screen may be non-display content, such as wallpaper or a black screen without actual content. The data visualization screen may also not include the non-preset view area, that is, the preset view area is set in full screen.

[0072] Furthermore, the preset view area may be divided into a plurality of sub-view areas, which may be divided evenly, such as into 3×3 sub-view areas, a total of nine sub-view areas, or may be divided into a plurality of sub-view areas of different sizes according to actual needs.

[0073] Each subview area is used to display one or more data tasks in the subsequent steps. Before displaying these data tasks in the subsequent steps, it is necessary to analyze these data tasks and their task information so as to formulate a reasonable differentiated display strategy to meet the different usage requirements of users.

[0074] In a preferred embodiment, the task information also includes data density. The clustering of the data tasks and building a tree index structure according to the clustering results includes:

[0075] A hierarchical clustering algorithm is used to perform preliminary clustering on each of the data tasks; wherein, during the preliminary clustering process, the Ward link method is used to minimize the variance of the clustering clusters; data tasks whose data density is less than or equal to a preset density threshold in the preliminary clustering results are eliminated; the DBSCAN algorithm is used to refine the clustering clusters corresponding to the eliminated data, and the ε parameter and domain density threshold of the clustering cluster are adjusted to obtain an optimized clustering result; based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

[0076] Exemplarily, a hierarchical clustering algorithm is first used to perform preliminary clustering on each of the data tasks. Specifically, the Ward link method is selected to minimize the variance within the cluster, and the number of clusters is set to 10, so that a hierarchical structure containing 10 main clusters can be initially formed. Further, considering that there are many data tasks that actually need to be displayed in some scenarios, the data tasks with a data density less than or equal to the preset density threshold are eliminated in the preliminary clustering results, which can remove a part of the noise and eliminate unimportant or useless data tasks, so as to selectively display them. If these data tasks need to be displayed later, they can be added to the queue to be displayed in the later stage of the display. Then, the DBSCAN algorithm is used to further refine these eliminated clusters, setting the ε parameter to 5 and the domain density threshold MinPts to 5 to adapt to the density distribution of the data, which helps to identify noise points with fewer edges and clusters of non-standard shapes. In this way, complex distributions in high-dimensional data can be processed to ensure that each cluster reflects the intrinsic structure of the data more accurately, so that a multi-level tree index structure can be constructed based on the optimized clustering results.

[0077] Step S2: determining the priority of each of the data tasks according to the tree index structure.

[0078] In this embodiment, the tree index structure can be updated in real time according to the importance of the data task.

[0079] Exemplarily, the method for updating the tree index structure includes:

[0080] Calculate the data volume, statistical indicators and task type of the data task corresponding to each node; wherein the statistical indicators are obtained by analyzing the optimized clustering results; monitor the tree index structure, and when at least one of the data volume, statistical indicators and task type of the data task corresponding to the lower-level node changes, update at least one of the data volume, statistical indicators and task type of the data task corresponding to the upper-level node through incremental calculation; wherein, before the incremental calculation update, it also includes: when the format type of the data task corresponding to the lower-level node is inconsistent with that of the data task corresponding to the upper-level node, convert the format type of the data task corresponding to the upper-level node according to the format type of the data task corresponding to the lower-level node.

[0081] Furthermore, the data visualization screen of this embodiment is often aimed at comprehensive data tasks, that is, the data of multiple subtasks are analyzed and counted to obtain the above-mentioned comprehensive data tasks. During the update process, for data tasks or nodes of tree index structures, when presenting data of specific statistical types, the data tasks to be displayed can also monitor their statistical indicators, such as the total number, mean and variance of some task parameters. At this time, when performing incremental updates, the updated upper node mean and variance are calculated by weighted average based on the mean and variance of the changed data of the lower nodes.

[0082] When updating the upper node index, if a new lower node is added, the statistical index of the new node is aggregated with the original upper node index to obtain the updated upper node index. If the lower node needs to be deleted, the statistical index of the deleted node is subtracted from the original upper node index to obtain the updated upper node index. As in the above example, the above incremental update steps are recursively executed from bottom to top until the root node index is updated, that is, the statistical index update of the entire tree structure is completed without retraversing and calculating the data of the entire tree.

[0083] Furthermore, determining the priority of each data task based on the tree index structure includes: determining the hierarchical relationship of the tree index structure based on the root node of the tree index structure; calculating the cosine similarity between each node of the tree index structure and the layer it belongs to based on the hierarchical relationship; determining the importance of each node based on the order of each cosine similarity from large to small, thereby determining the priority of each data task in turn.

[0084] Exemplarily, based on the root node of the tree index structure and the multiple paths extending from it, it is possible to determine how many layers the tree index structure has, and then based on the similarity between the node and the layer it is in, the importance (or relevance) of each node can be determined, and the priority of each data task can be determined using the similarity, and the data tasks can be sorted in order from highest to lowest priority, thereby determining the order of subsequent loading of the data tasks.

[0085] Step S3, obtaining the current zoom level of each sub-view area; and determining the data granularity of the data to be displayed in each sub-view area according to each current zoom level.

[0086] In this step, each sub-view area may have a preset independent zoom level, that is, the zoom levels of the sub-view areas may be the same or different.

[0087] For example, when the zoom level of the current subview area is 1, the data granularity of the data (task) to be displayed corresponding to the subview area can be selected as "county level". When the zoom level is 10, the corresponding data granularity can be "city level". When the zoom level is 100, the corresponding data granularity can be "provincial level".

[0088] The above examples of data granularity are only used as illustrative illustrations to illustrate the technical solution of this application. In addition, depending on the type and content of the data task, it may not be limited to the division of "province", "city" or "county", but may also be divided into another dimension such as "meter", "decimeter", "centimeter" or "millimeter".

[0089] Step S4, according to each of the data granularities, in each of the sub-view areas, a progressive data loading strategy is adopted and combined with the priority of each of the data tasks to sequentially load each of the data tasks.

[0090] In this step, as a preferred implementation, according to each data granularity, in each sub-view area, a progressive data loading strategy is adopted and combined with the priority of each data task to sequentially load each data task, including:

[0091] Construct multiple loading threads, each loading thread corresponds to a sub-view area; transmit each of the data tasks to each loading thread in sequence according to the priority; obtain the data block size threshold corresponding to each of the loading threads; when the total size of each data task in the loading thread exceeds the data block size threshold, divide all data tasks in the loading thread into several data sets (no classification is required when the total size does not exceed the data block size threshold); adopt a progressive data loading strategy, and load each of the data sets in sequence according to the corresponding data granularity in each of the sub-view areas; wherein data tasks of the same data set are loaded in parallel.

[0092] In this preferred embodiment, a separate loading thread is constructed for each subview area, and a shared memory mechanism can be established between loading threads to store data obtained by the loading threads, avoiding frequent data copy operations and reducing inter-thread communication overhead. Multiple loading threads can perform loading operations in parallel to improve overall data loading efficiency.

[0093] Furthermore, each of the data tasks is transmitted to each loading thread in sequence according to the priority; the data block size threshold corresponding to each of the loading threads is obtained, and when the total size of each data task of the same loading thread exceeds the data block size threshold, for example, 10MB, the data tasks in the loading thread can be classified into several data sets (for example, n), each of which contains one or more data tasks. Among them, except for the last data set, the first n-1 data sets need to be less than or equal to 10MB, and the last data set is the total size of each data task of the loading thread minus 10×(n-1)MB. Furthermore, for the same loading thread, all data tasks in the same data set can be loaded at the same time, that is, loaded in units of data sets in sequence according to priority and data granularity, so as to avoid excessive pressure on the loading thread, while well balancing the loading speed and improving loading efficiency.

[0094] Step S5, displaying the loaded data tasks according to the task coordinates.

[0095] In this step, for each data task, the task coordinates can be preset, that is, the location where the data task is to be displayed. Each data task is displayed in a corresponding sub-view area. In this embodiment, the data tasks can be displayed in sequence according to the order in which they are loaded.

[0096] When a user performs an interactive operation on the data visualization screen, this embodiment responds to the interactive operation on the data visualization screen; uses a pre-configured data mapping table to query according to the interactive operation, determines the type of interactive event, and determines whether the data view needs to be updated based on the type of interactive event; when it is determined that an update is required (for example, when the user queries the interactive event type of "zoom" through the interactive operation of "mouse click"), the corresponding target interactive task that needs to be updated (for example, the specific update content is the above-mentioned "zoom") is determined based on the effective area of ​​the interactive operation on the data visualization screen and the task coordinates; according to the interactive event type, the corresponding data range and data granularity are extracted; through the asynchronous update interface, the range parameters of the target interactive task are updated using the extracted data range (for example, partial update or full update), and the updated target interactive task is displayed based on the extracted data granularity (for example, the data granularity is scaled from "provincial level" to "municipal level" or "county level").

[0097] Correspondingly, such as Figure 2 As shown, the present invention application also provides a data visualization large screen implementation system, wherein the display content of the data visualization large screen includes a preset view area, and the preset view area includes multiple sub-view areas;

[0098] The implementation system includes a construction module 201, a priority determination module 202, a granularity determination module 203, a loading module 204 and a display module 205; wherein,

[0099] The construction module 201 is used to obtain multiple data tasks to be displayed on the data visualization screen and task information of each data task; cluster each data task and construct a tree index structure according to the clustering result; wherein the task information includes task coordinates;

[0100] The priority determination module 202 is used to determine the priority of each of the data tasks according to the tree index structure;

[0101] The granularity determination module 203 is used to obtain the current zoom level of each sub-view area; and determine the data granularity of the data to be displayed in each sub-view area according to each current zoom level;

[0102] The loading module 204 is used to load each of the data tasks in sequence according to each of the data granularities and in each of the sub-view areas by adopting a progressive data loading strategy and combining the priority of each of the data tasks;

[0103] The display module 205 is used to display the loaded data tasks according to the task coordinates.

[0104] As a preferred solution, the task information also includes data density; the construction module 201 clusters each of the data tasks and constructs a tree index structure according to the clustering results, including:

[0105] The construction module 201 performs preliminary clustering on each of the data tasks by using a hierarchical clustering algorithm; wherein, in the process of the preliminary clustering, the Ward link method is used to minimize the variance of the clusters;

[0106] Eliminate data tasks whose data density is less than or equal to the preset density threshold in the preliminary clustering results;

[0107] The DBSCAN algorithm is used to refine the clusters corresponding to the removed data, and the ε parameter and domain density threshold of the clusters are adjusted to obtain the optimized clustering results;

[0108] Based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

[0109] As a preferred solution, the method for updating the tree index structure includes:

[0110] Calculating the data volume, statistical indicators and task type of the data task corresponding to each node; wherein the statistical indicators are obtained by analyzing the optimized clustering results;

[0111] Monitor the tree index structure, and when at least one of the data volume, statistical indicators, and task type of the data task corresponding to the lower-level node changes, update at least one of the data volume, statistical indicators, and task type of the data task corresponding to the upper-level node through incremental calculation;

[0112] Among them, before the incremental calculation update, it also includes: when the format type of the data task corresponding to the lower-level node is inconsistent with the format type of the data task corresponding to the upper-level node, the format type of the data task corresponding to the upper-level node is converted according to the format type of the data task corresponding to the lower-level node.

[0113] As a preferred solution, the priority determination module 202 determines the priority of each of the data tasks according to the tree index structure, including:

[0114] The priority determination module 202 determines the hierarchical relationship of the tree index structure according to the root node of the tree index structure;

[0115] According to the hierarchical relationship, respectively calculating the cosine similarity between each node of the tree index structure and the layer in which it is located;

[0116] According to the order of cosine similarities from large to small, the importance of each node is determined, thereby determining the priority of each data task in turn.

[0117] As a preferred solution, the implementation system further includes an update module; the update module is used to:

[0118] Responding to interactive operations on the data visualization screen;

[0119] Using a pre-configured data mapping table, querying according to the interaction operation to determine the interaction event type;

[0120] Determine the target interactive task according to the action area of ​​the interactive operation on the data visualization screen and the task coordinates; extract the corresponding data range and data granularity according to the interactive event type;

[0121] The range parameter of the target interactive task is updated by using the extracted data range through an asynchronous update interface, and the updated target interactive task is displayed according to the extracted data granularity.

[0122] As a preferred solution, the loading module 204 adopts a progressive data loading strategy in each sub-view area according to each data granularity and combines the priority of each data task to sequentially load each data task, including:

[0123] The loading module 204 constructs multiple loading threads, each loading thread corresponds to a sub-view area;

[0124] Transmitting each of the data tasks to each loading thread in sequence according to the priority;

[0125] Obtaining a data block size threshold corresponding to each of the loading threads;

[0126] When the total size of each data task in the loading thread exceeds the data block size threshold, dividing all data tasks in the loading thread into a plurality of data sets;

[0127] A progressive data loading strategy is adopted, and in each sub-view area, each data set is loaded in sequence according to the corresponding data granularity; wherein data tasks of the same data set are loaded in parallel.

[0128] Compared with the prior art, the present invention has the following beneficial effects:

[0129] An embodiment of the present invention provides a method and system for implementing a large data visualization screen, wherein the display content of the large data visualization screen includes a preset view area, and the preset view area includes multiple sub-view areas; the implementation method includes: obtaining multiple data tasks to be displayed on the data visualization screen and task information of each data task; clustering each data task, and constructing a tree index structure according to the clustering results; wherein the task information includes task coordinates; determining the priority of each data task according to the tree index structure; obtaining the current zoom level of each sub-view area; according to each current zoom level, determining the data granularity of the data to be displayed in each sub-view area; according to each data granularity, in each sub-view area, adopting a progressive data loading strategy and combining the priority of each data task, loading each data task in turn; according to each task coordinate, displaying the loaded data task. The present invention obtains the current zoom level of the sub-view area to determine the data granularity of the data to be displayed in the current sub-view area, which can meet the differentiated needs of users for different sub-view areas and different tasks, improve the intelligence level of data visualization large-screen data display, and thus meet the needs of flexible and complex application scenarios, and has universal applicability; in addition, by clustering multiple data tasks to be displayed and constructing a tree index structure to determine the priority of the data task, and loading and displaying each data task in turn according to the priority, orderly loading and display of data tasks is achieved, which can further meet the user's display needs for data tasks of different priorities and further enhance the user experience.

[0130] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for realizing a data visualization large screen, characterized in that: The display content of the data visualization large screen includes a preset view area, and the preset view area includes a plurality of sub-view areas; The implementation method includes: Acquire multiple data tasks to be displayed on the data visualization screen and task information of each data task; cluster each data task and construct a tree index structure according to the clustering results; wherein the task information includes task coordinates; Determining the priority of each of the data tasks according to the tree index structure; Acquire the current zoom level of each sub-view area; and determine the data granularity of the data to be displayed in each sub-view area according to each current zoom level; According to each of the data granularities, in each of the sub-view areas, a progressive data loading strategy is adopted and combined with the priority of each of the data tasks to sequentially load each of the data tasks; specifically comprising: constructing multiple loading threads, each loading thread corresponds to a sub-view area; transmitting each of the data tasks to each of the loading threads in sequence according to the priority; obtaining a data block size threshold corresponding to each of the loading threads; when the total size of each of the data tasks in the loading thread exceeds the data block size threshold, dividing all of the data tasks in the loading thread into a number of data sets; adopting a progressive data loading strategy, in each of the sub-view areas, sequentially loading each of the data sets according to the corresponding data granularity; wherein data tasks of the same data set are loaded in parallel; According to the coordinates of each task, the loaded data tasks are displayed.

2. The method for realizing a data visualization large screen according to claim 1, characterized in that: The task information also includes data density; clustering the data tasks and constructing a tree index structure according to the clustering results, including: Performing preliminary clustering of each of the data tasks by a hierarchical clustering algorithm; wherein, in the process of preliminary clustering, the Ward link method is used to minimize the variance of the clustering clusters; Eliminate data tasks whose data density is less than or equal to the preset density threshold in the preliminary clustering results; The DBSCAN algorithm is used to refine the clusters corresponding to the removed data, and the ε parameter and domain density threshold of the clusters are adjusted to obtain the optimized clustering results; Based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

3. The method for realizing a data visualization large screen according to claim 2, characterized in that: Determining the priority of each of the data tasks according to the tree index structure includes: Determining a hierarchical relationship of the tree index structure according to a root node of the tree index structure; According to the hierarchical relationship, respectively calculating the cosine similarity between each node of the tree index structure and the layer in which it is located; According to the order of cosine similarities from large to small, the importance of each node is determined, thereby determining the priority of each data task in turn.

4. The method for realizing a data visualization large screen according to claim 1, characterized in that: The implementation method also includes: Responding to interactive operations on the data visualization screen; Using a pre-configured data mapping table, querying according to the interaction operation to determine the interaction event type; Determine the target interactive task according to the action area of ​​the interactive operation on the data visualization screen and the task coordinates; extract the corresponding data range and data granularity according to the interactive event type; The range parameter of the target interactive task is updated by using the extracted data range through an asynchronous update interface, and the updated target interactive task is displayed according to the extracted data granularity.

5. A data visualization large screen implementation system, characterized in that: The display content of the data visualization large screen includes a preset view area, and the preset view area includes a plurality of sub-view areas; The implementation system includes a construction module, a priority determination module, a granularity determination module, a loading module and a display module; wherein, The construction module is used to obtain multiple data tasks to be displayed on the data visualization screen and task information of each data task; cluster each data task and construct a tree index structure according to the clustering results; wherein the task information includes task coordinates; The priority determination module is used to determine the priority of each of the data tasks according to the tree index structure; The granularity determination module is used to obtain the current zoom level of each sub-view area; and determine the data granularity of the data to be displayed in each sub-view area according to each current zoom level; The loading module is used to load each data task in sequence in each sub-view area according to each data granularity, using a progressive data loading strategy and combining the priority of each data task; specifically comprising: constructing multiple loading threads, each loading thread corresponding to a sub-view area; transmitting each data task to each loading thread in sequence according to the priority; obtaining a data block size threshold corresponding to each loading thread; when the total size of each data task in the loading thread exceeds the data block size threshold, dividing all data tasks in the loading thread into several data sets; using a progressive data loading strategy, in each sub-view area, sequentially loading each data set according to the corresponding data granularity; wherein data tasks of the same data set are loaded in parallel; The display module is used to display the loaded data tasks according to the task coordinates.

6. A data visualization large screen implementation system as claimed in claim 5, characterized in that: The task information also includes data density; The construction module clusters each of the data tasks and constructs a tree index structure according to the clustering results, including: The construction module performs preliminary clustering of each of the data tasks through a hierarchical clustering algorithm; wherein, during the preliminary clustering process, the Ward link method is used to minimize the variance of the clusters; Eliminate data tasks whose data density is less than or equal to the preset density threshold in the preliminary clustering results; The DBSCAN algorithm is used to refine the clusters corresponding to the removed data, and the ε parameter and domain density threshold of the clusters are adjusted to obtain the optimized clustering results; Based on the optimized clustering result, a multi-level tree index structure is constructed; wherein each node of the tree index structure corresponds to a data task of the optimized clustering result, and the link relationship between the nodes corresponds to the hierarchical relationship between clusters.

7. A data visualization large screen implementation system as claimed in claim 6, characterized in that: The priority determination module determines the priority of each of the data tasks according to the tree index structure, including: The priority determination module determines the hierarchical relationship of the tree index structure according to the root node of the tree index structure; According to the hierarchical relationship, respectively calculating the cosine similarity between each node of the tree index structure and the layer in which it is located; According to the order of cosine similarities from large to small, the importance of each node is determined, thereby determining the priority of each data task in turn.

8. A data visualization large screen implementation system as claimed in claim 5, characterized in that: The implementation system further includes an update module; the update module is used to: Responding to interactive operations on the data visualization screen; Using a pre-configured data mapping table, querying according to the interaction operation to determine the interaction event type; Determining a target interactive task according to an active area of ​​the interactive operation on the data visualization screen and the task coordinates; According to the interaction event type, extract the corresponding data range and data granularity; The range parameter of the target interactive task is updated by using the extracted data range through an asynchronous update interface, and the updated target interactive task is displayed according to the extracted data granularity.

Images