An atlas visual query system

Abstract

The application relates to a map, namely, a visual query system, which comprises the following function modules: a visual map screening function module, which is used for screening graph data on a map page; an edge query function module, which is used for querying according to the type of an edge; a path and snapshot saving function module, which is used for saving a current path and results in an operation path; a name display function module, which is used for setting the display type of a name; a statistical function module, which is used for statistically describing graph data; a page turning function module, which is used for turning the current graph data page; a permission management function module, which is used for managing permissions; an early warning function module, which is used for realizing the early warning function of graph data abnormality according to the hop check-off method; a label real-time dynamic function module, which is used for realizing the real-time dynamic change of a label; and a map generation time prediction function module, which is used for providing the progress estimation of map generation. Compared with the prior art, the application has the advantages of optimizing visual experience, improving abnormal early warning capability and the like.

Description

Technical Field

[0001] This invention relates to a query system, and more particularly to a map-based visual query system. Background Technology

[0002] In my country, knowledge graphs are an independent discipline under the category of Artificial Intelligence. As an engineering discipline, the specific methods of knowledge graphs largely originate from other fields of computer science or artificial intelligence. Knowledge graphs are an application-oriented discipline aimed at solving complex real-world problems. The key to the quality of the final product lies in the selection and strategy; it requires considering which method to use to solve the problem under current technological conditions.

[0003] Existing technologies, such as data analysis tools like Excel, SAS, and DB2, possess powerful data analysis and dynamic graphical display capabilities. However, their data structures are two-dimensional (row and column), lacking the intuitive graph features (such as path length) of graph data structures. Therefore, calculating the transaction relationship path between two customers is more computationally intensive than with graph data structures. Chinese patent CN 202210168248.3 discloses a knowledge graph analysis system and method, employing a hybrid storage technology of MySQL and Neo4j, solving the problems of traditional databases' inability to flexibly change entity and relationship attributes and low data query efficiency. However, it only considers the data analysis and display of the knowledge graph, without mining the data to form early warning rules. Furthermore, the existing graph database Neo4j is more specialized in graph databases, and its core market competitiveness lies mainly in improving data performance. Therefore, its application in data analysis scenarios still needs improvement. Refining scenario details typically requires extensive industry experience. A mismatch between the problem and the solution inevitably reduces the relevance of the solution.

[0004] Therefore, how to integrate the fields of data analysis and graph databases / knowledge graphs, and leverage the advantages of graphs in the field of data analysis, has become a technical problem that needs to be solved. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a map-based visual query system.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A map-based instant query system, the system comprising:

[0008] The map and graph filtering module is used to filter, query, and display map data on the map and graph page.

[0009] The edge query module is used to perform statistical, query, and filtering operations based on edge type.

[0010] A save path and snapshot function module is configured to record the current path and result in the operation path, so as to realize continuous development next time.

[0011] A page turning function module is configured to realize the page turning function for the current graph data page.

[0012] A pre-warning function module is configured to realize the pre-warning function when the graph data is abnormal according to the checkers peeling method.

[0013] Further, the view spectrum filtering function module has the following specific logic:

[0014] S101, setting the advanced query condition in the search toolbar of the current graph display interface and clicking the query;

[0015] S102, updating the current graph dynamically to the new graph corresponding to the advanced query result data.

[0016] Further, the edge query function module has the following specific logic:

[0017] S201, obtaining the other point type on the edge according to the specified edge type and the main point, and if there are multiple edges, obtaining multiple point types;

[0018] S202, taking one point type of each point type as a group according to the point type, and forming multiple small groups;

[0019] S203, traversing the small groups, finding the main point from each point in the group, and recording the user point list; if the number range is included in the query condition, filtering the list according to the number range.

[0020] Further, the save path and snapshot function module has the following specific logic:

[0021] When the save path and snapshot function module is clicked, all data content and data state displayed on the current page are saved in the background database in the form of a json file; multiple snapshots are saved, when the snapshot list is opened and one of the snapshots is clicked, the system returns the data information corresponding to the snapshot in the database to the front end, and the front end renders the page according to the data content.

[0022] Further, the page turning function module has the following specific logic: whenever there is an operation of turning 5 pages forward or backward, a current turned number is recorded for the target point, and when the page is turned again, the page is turned according to the recorded value, and when the point disappears due to the undo or canvas clearing operation, the recorded turned number on the point is also deleted synchronously.

[0023] Further, the checkers peeling method has the following specific logic:

[0024] S501, specify the target node type, specify the node type or edge type for jumping, wherein the target node type cannot be set as the jumping node type at the same time, execute S502;

[0025] S502, establish a direct connection relationship between the target nodes of the target node type instead of the original jumping path, form a new graph structure, execute S503, between the nodes reachable by the node type or edge type for jumping;

[0026] S503, set a flag bit for each node of the target node type as the number of single-step reachable nodes, determine whether it is an isolated point or an isolated chain, wherein the flag bit of the isolated point is 0, the flag bit of the end node of the isolated chain is 1, and the flag bit of the remaining nodes is 2, execute S504;

[0027] S504, remove the isolated points and the end nodes of the isolated chains according to the flag bits, and re-calculate whether the flag bit of the previous node of the end node is still 1 or 0, and remove the isolated chains in a loop until the flag bits of the remaining nodes are all 2, form a loop diagram, execute S505;

[0028] S505, clear the flag bits of each node remaining in the loop diagram, and give the order number of the node traversal operation according to the rule, jump step by step, and peel off step by step through the jumping path after node orientation and recording, execute S506;

[0029] S506, start jumping from two target node type nodes, increase the jumping step length by 1 step by step, and iteratively execute S507-S509 until there is no new jumping node, or the jumping step length reaches the preset N steps, and N steps or more are included in the ring above N order;

[0030] S507, the rule of jumping: the next jumpable node cannot be the starting node, and cannot be a node already on the jumping path; if a new node can be found, the jump can continue, and the flag bit is set to 1, and the jump is continued when S506 is iterated; if there is no new node that can continue to jump, the flag bit is set to 0, indicating that the path is a closed loop or a ring in a ring, return to the previous node, traverse other branches until the starting node is returned, and there is still no new node that can continue to jump, then peel off as an N+1 edge independent ring;

[0031] S508, save the currently traversed jumping path and the flag bit;

[0032] S509, peel off the nodes with flag bit 0 to form an N+1 edge independent ring.

[0033] Further, the system further comprises:

[0034] The name display function module is used to set the display type of the name as needed; the types include full name, name abbreviation, name hiding, and name encryption;

[0035] Statistical function module for statistical description of graph data, the content of statistical description including mode, median and data type;

[0036] Authority management function module for managing user organization authority, post authority and individual user authority;

[0037] Label real-time dynamic function module for realizing real-time dynamic change of labels;

[0038] Graph generation time prediction function module for providing progress estimation of graph generation.

[0039] Further, the name display function module is used to meet the information desensitization requirements of supervision while retaining the associated contact required for data analysis, and specifically includes the following display types:

[0040] Full name: traverse each point in the page, get the name attribute of the point, and directly assign the name attribute to the display attribute;

[0041] Abbreviation: traverse each point in the page, get the name attribute of the point, and set the text exceeding the fixed number of characters to be replaced by an ellipsis;

[0042] Name hidden: traverse each point in the page, and assign the value of the display attribute to an empty string;

[0043] Name encryption: when the attribute of a point or edge in the graph is an ID card or a name, execute encryption logic to encrypt and display the characters at a fixed position in the string.

[0044] Further, the specific logic of the statistical function module is:

[0045] Traverse all points of the selected point type, calculate the name of the point, and statistically describe the information of the graph data;

[0046] The specific logic of the authority management function module is:

[0047] Add a page-invisible attribute field to the attributes of points and edges, which contains permission information. When the page queries graph data, the background will add a filtering condition for the permission attribute field, thereby controlling the user-visible information at the data source end;

[0048] The specific logic of the label real-time dynamic function module is:

[0049] Add a label column to the data structure, real-time calculate and reset the label value according to the preset label conversion rule, and display the corresponding graph after resetting the label on the page.

[0050] Further, the atlas generation time prediction function module has the following specific logic:

[0051] The database information and graph processing information are read through the interface, the graph progress is estimated by combining the query statement, table level and query condition complexity, and when the graph is not displayed within 5 seconds, the graph progress estimation is provided, and the user decides whether to continue waiting or change the query strategy according to the progress estimation.

[0052] Compared with the prior art, the present application has the following beneficial effects:

[0053] First, the atlas of the present application is an atlas system, which integrates the knowledge graph field of the graph database into the abnormal early warning application scenario in the data analysis field from the perspective of the graph dimension, can mine abnormal risk points related to graph features, sets corresponding early warning rules, reduces the path of case occurrence, and at the same time, the 4D graph data immersive visual data analysis experience can also stimulate the user's intuitive inspiration.

[0054] Second, the present application realizes the recording and saving of user operations through the saving snapshot function, avoids the passive irreproducible of previous analysis content due to the update of basic data when logging in next time, and through the sharing path (operation log), the data analysis extraction process can also be conveniently and clearly and quickly standardized shared with other users.

[0055] Third, the present application realizes the more perfect display of the atlas through the design of multiple function modules, and meets various needs of users. DETAILED DESCRIPTION

[0056] Figure 1 The system module schematic diagram of the present application is shown in the figure;

[0057] Figure 2 The implementation logic schematic diagram of the visual atlas screening function module of the present application is shown in the figure;

[0058] Figure 3 The implementation logic schematic diagram of the edge query function module of the present application is shown in the figure;

[0059] Figure 4 The implementation logic schematic diagram of the saving path and snapshot function module of the present application is shown in the figure;

[0060] Figure 5 The implementation logic schematic diagram of the page turning function module of the present application is shown in the figure;

[0061] Figure 6 The implementation logic schematic diagram of the early warning function module of the present application is shown in the figure;

[0062] Figure 7 The display type schematic diagram of the name display function module of the present application is shown in the figure;

[0063] Figure 8Logic diagram for realizing the statistical function module of the application;

[0064] Figure 9 Logic diagram for realizing the authority management function module of the application;

[0065] Figure 10 Logic diagram for realizing the label real-time dynamic function module of the application;

[0066] Figure 11 Logic diagram for realizing the atlas generation time prediction function module of the application. DETAILED DESCRIPTION

[0067] The application will be described in detail below in combination with the drawings and specific embodiments. The embodiments are implemented on the premise of the technical solution of the application, and detailed implementation modes and specific operation processes are given, but the protection scope of the application is not limited to the following embodiments.

[0068] EMBODIMENT

[0069] As shown in the drawings, Figure 1 An atlas visual query system includes an atlas visual filtering function module, an edge-based query module, a path and snapshot saving function module, a name display function module, a statistical function module, a page turning function module, an authority management function module, a warning function module, a label real-time dynamic function module, and an atlas generation time prediction function module.

[0070] The atlas visual filtering function module does not need to exit the atlas display interface and go to a traditional two-dimensional table query page to complete data filtering, and then return to the atlas display interface to view the atlas display effect corresponding to the two-dimensional result data, unlike the traditional process. This module can realize direct data filtering on the atlas page, thus providing a smoother immersive visual experience and stimulating user data intuition and inspiration. In the traditional process of atlas filtering query, the atlas is only used for direct display of result data, and the query is completed through a traditional report (a relational database) with a non-graph structure. The traditional atlas filtering process is as follows:

[0071] 1. Close the current atlas display interface

[0072] 2. Return to the traditional report (a two-dimensional table, a relational database) filtering page

[0073] 3. Complete data preparation work after data filtering

[0074] 4. Return to the atlas display interface to display the data result of step 3 in the structure of graph data.

[0075] As shown in the drawings, Figure 2As shown, the view spectrum filtering function module does not need to jump out of the current spectrum page, and does not need the traditional filtering page of the two-dimensional table. It is a true image retrieval. The logic of the view spectrum filtering function module is as follows:

[0076] S101, set advanced query conditions in the search toolbar of the current spectrum display interface (reference excel function);

[0077] S102, the current spectrum is dynamically updated to the spectrum corresponding to the result data after advanced filtering.

[0078] The query coherence experience of the spectrum view query is beneficial to the user to feel the repetition and difference in the image, improve the accuracy of intuitive prediction, and adjust the prediction of the early warning rule through rapid verification to strengthen the accuracy of intuitive prediction. There is no need to find the background personalized analysis of the number of personnel, or to wait for several weeks of fixed table header report function development online.

[0079] The traditional method of spectrum query is to expand from a point to find associated edges and points. This traditional method only needs to perform a one-layer select query from the point to obtain the result. The edge query function module specified by the application, by specifying the main point type and the edge type to be expanded, can group the specified main point type in the dimension of the edge type. For example, the graph data has "mobile phone point, contact way edge, user point", and after specifying the user and the contact way, the total number of "user points" connected by "contact way type edge" to each "mobile phone point" can be calculated. In addition, this module also supports the specification of multiple edge types, which can realize the search for users or user numbers with the same mobile phone number and the same address. Figure 3 As shown, the specific logic is as follows:

[0080] S201, according to the specified edge type and the main point, obtain another point type on the edge. If it is multiple edges, multiple point types will be obtained.

[0081] S202, according to the point type, take one of each point type as a group, and many small groups will be formed.

[0082] S203, traverse the group, find the main point from each point in the group, and record the user point list; if there is a quantity range in the query condition, the user point list is filtered according to the quantity range.

[0083] As shown, Figure 4As shown, the saving path and snapshot function module records user operations in the operation path (including undo operations), supports path saving and snapshot saving of the current result, and users can log in multiple times to realize continuous development, while also avoiding the passive irreproducibility of previous analysis content due to changes in the underlying data update when logging in next time. Through sharing paths (operation logs), the data analysis extraction process can also be conveniently and clearly and quickly standardized shared with other users. The specific implementation logic is that each operation code of the graph exploration page has log recording logic, and when the user operates on the page, each step of operation will record the name of the operation and the data of the operation (such as the object of the operation and other information) in the foreground memory, and all operation steps can be viewed at any time through the page button function. When the user clicks the snapshot saving function, all data content and data state (such as hidden points, virtual points, etc.) displayed on the current page will be saved in a json file in the background database. Multiple snapshots can be saved, and when the user opens the snapshot list and clicks one of the snapshots, the system will return the data information corresponding to this snapshot in the database to the foreground, and the foreground will render the page according to the data content.

[0084] The page turning function module provides page turning function, which is convenient for observing the abnormal nodes of concentration more clearly; and the page turning function module also supports single-layer multi-selection expansion of specified "relationship type" or "object type". For example, Figure 5 As shown, the specific implementation logic of the page turning function module is that each time there is an operation of turning 5 pages forward or backward, a current number of turned pages is recorded for the target point, and when the point disappears due to undo or canvas clearing operation, the recorded number of turned pages for the point is also deleted synchronously.

[0085] As shown, Figure 6 The early warning function module is used to realize the abnormal early warning function of graph data according to the checkers peeling method; and the checkers peeling method specifically includes:

[0086] S501, specifying a target node type, specifying a node type or edge type for jumping, wherein the target node type cannot be set as the jumping node type at the same time, and S502 is executed;

[0087] S502, establishing a direct connection relationship between target nodes instead of the original jumping path between the target nodes of the node type reached by the node type or edge type for jumping, forming a new graph structure, and S503 is executed;

[0088] S503, setting a flag bit for each target node type node as the number of single-step reachable nodes, and determining whether it is an isolated point or an isolated chain, wherein the flag bit of the isolated point is 0, the flag bit of the terminal node of the isolated chain is 1, and the flag bits of the remaining nodes are 2, and S504 is executed;

[0089] S504, according to the flag, remove isolated points and isolated chain end nodes, and re-calculate the last node flag of the end node whether it is still 1 or 0, and remove the isolated chain in a loop until the remaining node flags are all 2, forming a ring diagram, and executing S505;

[0090] S505, clear the flags of each node remaining in the ring diagram, and give the order number of the node traversal operation according to the rule, and gradually jump, peel off through the jump path after the node orientation and recording, and execute S506;

[0091] S506, starting from the jump 2 target node type node, the jump step is gradually increased by 1, and S507-S509 are iteratively executed until there is no new jump node, or the jump step reaches the preset N steps, and N steps above are returned to the ring above N;

[0092] S507, the rule of jumping: the next jumpable node cannot be the starting node, and cannot be a node already on the jump path; if a new node can be found, the jump can continue, and the flag is set to 1, and the jump is continued when S506 is iterated; if there is no new node that can continue to jump, the flag is set to 0, indicating that the path is a closed loop or a ring in a ring, returning to the last node and traversing other branches until returning to the starting node, and still no new node can continue to jump, then peel off as an N+1 edge independent ring;

[0093] S508, save the path of the current traversed jump and the flag;

[0094] S509, peel off the node with flag 0 to form an N+1 edge independent ring.

[0095] Taking the traffic bank's graph mining early warning rule as an example, the traffic bank's customer nodes and telephone nodes, as well as "guarantee relationship", "telephone relationship of credit users", and "telephone relationship of guarantee users", are displayed using a graph.

[0096] First, specify "(customer) type node", which can jump "telephone node"; then remove "telephone node", and then mark isolated points, chains, isolated rings, etc. through flags; finally, peel off the chain and three-edge isolated ring, and the remaining final high-order complex ring-in-ring, as shown in Figure 8 .

[0097] Set the early warning rule of customer risk concentration degree abnormally high and complex, such as guarantee social circle complex early warning: there is a high-order guarantee ring above four edges in a certain traffic bank's public customer, and there is a complex ring characteristic of ring-in-ring. Further, the static parameter "four-edge ring" in the early warning rule can be proportioned and configured, such as "ring with edge number higher than 20 % of the ring".

[0098] Based on the effect of the diagram structure of the checkers peeling method, a local checkers peeling ring diagram can be realized by locking a node, that is, only the social circle of the locked node is simplified in a targeted manner, then the "edge" hiding function is used, and the structure is further split, so that the relationship structure is clearer, and the effect of highlighting the key nodes and key paths is achieved. The edges can also be given weights according to the guarantee amount, and the edge relationships with small weights and few associated nodes are automatically peeled off.

[0099] As Figure 7 shown, when the text content is too much, affecting the visual effect, the name display function module can set name full name, name abbreviation, name hiding and name encryption functions according to requirements, meet the information desensitization requirements of supervision while retaining the associated contact required for data analysis, as follows:

[0100] Name full name: traverse each point in the page, get the name attribute of the point, and directly assign the name attribute to the display attribute.

[0101] Name abbreviation: traverse each point in the page, get the name attribute of the point, and set the text exceeding the fixed number of characters to be replaced by...

[0102] Name hiding: traverse each edge in the page, and assign an empty string to the value of the display attribute.

[0103] Name encryption: When the attribute of a point or edge in the diagram is an ID card or a name, the encryption logic is executed, and the characters at a fixed position in the string are replaced with "*" characters.

[0104] The statistical function module provides basic statistical analysis functions such as statistical description, to facilitate users to set different data analysis filtering and processing strategies according to data characteristics. Clicking "contact address" can expand the data statistical description of "contact address", and clicking again can collapse it. As Figure 8 shown, the specific function implementation logic is as follows:

[0105] Traverse all points of the selected point type, calculate the name of the point, calculate the mode, median and other data, and display them on the page.

[0106] As Figure 9 shown, the permission management function module is used to manage the permissions of user institutions, post permissions and individual user permissions. The specific implementation logic is as follows: the system contains comprehensive permission control, from top to bottom, there are project level, graph level, point and edge level and attribute level. The permission control implementation of point and edge level is to add a page invisible attribute field in the attributes of point and edge, which contains permission information. When the page queries the graph data, the background will add a filtering condition for the permission attribute field, and then realize the control of the visible information of the user in the data source.

[0107] As Figure 10 shown, compared with the pre-processing data of static label, the label real-time dynamic function module is post-processing data, and the data interaction structure and mechanism are more complex, but it is more beneficial for customers to adjust the label rules in real time as needed. The specific implementation logic is as follows:

[0108] A label column is added to the data structure, and the label value is calculated and reset in real time according to the set label conversion rule. In this way, users can see the data and concentrate and refine key information as needed through label rules, reduce the number of nodes, and display more prominently.

[0109] For example, point <transaction TXN65496873287> is originally associated with point <transaction amount 39000>, and point <transaction TXN65496873288> is originally associated with point <transaction amount 49000>;

[0110] After label conversion, point <transaction TXN65496873287> and point <transaction TXN65496873288> are simultaneously associated with point <transaction amount interval label [2-5]>;

[0111] When the customer adjusts the label rule, the graph quickly responds, and point <transaction TXN65496873287> is originally associated with point <transaction amount interval label [2-4]>, and point <transaction TXN65496873288> is originally associated with point <transaction amount interval label [4-6]>. It is convenient for customers to adjust the collection view as needed and more effectively cluster and observe data.

[0112] As Figure 11 shown, the graph generation time prediction function module predicts the processing progress when the waiting time exceeds 5 seconds according to the size of the data set and the complexity of single operation, and the current resource state of the processor. If it is not completed in processing, it can stay at most at 99% progress. The specific implementation logic is as follows:

[0113] By reading the database information and graph processing information through the interface, combining the table, field number and table order of the query statement, and the complexity of the query condition, the graph generation time is predicted. When the graph is not displayed within 5 seconds, the progress estimation of the graph is provided, which is convenient for users to decide whether to continue to wait or change the query strategy.

[0114] The above detailed the preferred embodiments of the present application. It should be understood that those skilled in the art can make many modifications and changes without creative labor according to the concept of the present application. Therefore, any technical solution obtained by logical analysis, reasoning or limited experiment based on the existing technology according to the concept of the present application shall be within the protection scope determined by the claims.

Claims

1. A map-based instant query system, characterized in that, The system includes: The map and graph filtering module is used to filter, query, and display map data on the map and graph page. The edge query module is used to perform statistical, query, and filtering operations based on edge type. The save path and snapshot function module is used to record the current path and results in the operation path so that continuous development can be achieved on the next login; The page turning function module is used to enable page turning for the current graph data page; The early warning module is used to provide early warnings when graph data is abnormal, based on the checkers peeling method. The specific logic of the checkers peeling method is as follows: S501. Specify the target node type, specify the node type or edge type used for jumping. The target node type cannot be set to the jump node type at the same time. Execute S502. S502. Between target node types that can be reached through the node type or edge type used for jumping, establish direct connections between target nodes to replace the original jumping paths, forming a new graph structure, and execute S503. S503. Set the flag bit of each target node type to the number of nodes reachable in one step, and determine whether it is an isolated point or an isolated chain. The flag bit of an isolated point is 0, the flag bit of the end node of an isolated chain is 1, and the flag bit of the remaining nodes is 2. Execute S504. S504. Based on the flag bit, remove isolated points and the end nodes of isolated chains, and recalculate whether the flag bit of the previous node of the end node is still 1 or 0. Repeat the process of removing isolated chains until the flag bits of the remaining nodes are all 2, forming a loop diagram, and then execute S505. S505. Clear the flag bits of each node remaining in the ring diagram, and give the sequence number of the node traversal operation according to the rule. Through node orientation and recording of the subsequent jump path, jump step by step, peel off step by step, and execute S506. S506. Starting from two target node types, the jump step size is gradually increased by 1. S507-S509 are executed iteratively until there are no new jump nodes, or the jump step size reaches the preset N steps. Any jumps of more than N steps are classified into rings of order N or higher. S507, Jump Rules: The next jumpable node cannot be the starting node, nor can it be a node already on the jump path; if a new node can be found, the jump can continue, and the flag is set to 1, continuing the jump during the S506 iteration; if no new node can be found, the flag is set to 0, indicating that the path is a closed loop or a loop within a loop, return to the previous node, traverse other branches, until return to the starting node, and if no new node can be found, then it is stripped into an N+1 edge independent loop; S508. Save the currently traversed and jumped paths and their flags; S509. Strip the nodes with the flag bit set to 0 to form an N+1 edge independent loop.

2. The map-based instant query system according to claim 1, characterized in that, The specific logic of the instant view spectrum filtering function module is as follows: S101. Set advanced search criteria in the search toolbar of the current map display interface, and click Search; S102. The current map is dynamically updated to the new map corresponding to the advanced query result data.

3. The map-based instant query system according to claim 1, characterized in that, The specific logic of the edge-based query function module is as follows: S201. Based on the specified edge type and main point, obtain the other point type on the edge. If there are multiple edges, obtain multiple point types. S202. Based on point type, take one point of each point type as a group to form multiple groups; S203. Traverse the group, starting from each point in the group to find the main point, and record the list of user points; if the query condition includes a quantity range, then filter the list according to the quantity range.

4. The map-based instant query system according to claim 1, characterized in that, The specific logic of the save path and snapshot function module is as follows: When the save path and snapshot function modules are clicked, all data content and data status displayed on the current page will be saved as a JSON file in the backend database. Multiple snapshots can be saved. When the snapshot list is opened and one of the snapshots is clicked, the system will return the data information corresponding to this snapshot in the database to the front end, and the front end will render the page based on the data content.

5. The map-based instant query system according to claim 1, characterized in that, The specific logic of the page-turning function module is as follows: whenever there is an operation of turning forward or backward 5 pages, a current number of pages turned is recorded for the target point. When turning pages again, the page will be turned according to the recorded value. When a point disappears due to an undo or clear canvas operation, the number of pages turned recorded on the point is also deleted simultaneously.

6. The map-based instant query system according to claim 1, characterized in that, The system also includes: The name display function module is used to set the display type of the name as needed; the types include full name, name abbreviation, name hidden, and name encryption; The statistics module is used to provide statistical descriptions of graph data. The statistical descriptions include the mode, median, and data type. The access control module is used to manage user organization permissions, job permissions, and individual user permissions. The real-time dynamic labeling module is used to enable real-time dynamic changes to labels. The map generation time prediction module is used to provide a progress estimate for map generation.

7. A map-based instant query system according to claim 6, characterized in that, The aforementioned name display function module is used to meet regulatory requirements for information anonymization while retaining the relationships needed for data analysis, and specifically includes the following display types: Full Name: Traverse every point on the page, get the name attribute of the point, and directly assign the name attribute to the display attribute; Name Abbreviation: Traverse every point on the page, get the name attribute of the point, and set the text exceeding a fixed number of characters to be replaced with an ellipsis; Name hiding: Iterate through every point on the page and assign an empty string to the value of the displayed attribute; Name encryption: When the attribute of a point or edge in the graph is determined to be an ID card or a person's name, encryption logic is executed to encrypt and display characters at fixed positions in the string.

8. A map-based instant query system according to claim 6, characterized in that, The specific logic of the statistical function module is as follows: Iterate through all points of the selected point type, calculate the point names, and summarize the information of the descriptive graph data; The specific logic of the aforementioned permission management function module is as follows: Add invisible attribute fields to the properties of points and edges. These fields contain permission information. When the page queries graph data, the backend will add filtering conditions for the permission attribute fields, thereby enabling control over the information visible to users at the data source end. The specific logic of the real-time dynamic tag function module is as follows: The data structure is equipped with a label column. According to the set label conversion rules, the label values ​​are calculated and reset in real time, and the corresponding graphs after the labels are reset are displayed on the page.

9. A map-based instant query system according to claim 6, characterized in that, The specific logic of the map generation time prediction function module is as follows: The system reads database and graph processing information through an interface, and combines the query statement, table size, and query condition complexity to estimate the graph progress. If the graph is not displayed within 5 seconds, the system provides a progress estimate, which the user can then use to decide whether to continue waiting or change their query strategy.

Images