Information interception method and device based on graph calculation, equipment and readable storage medium

Abstract

The application provides an information interception method and device based on graph calculation, equipment and a readable storage medium, comprising: obtaining current business data, the current business data being the type and attribute value of an entity related to a business being applied in an application scenario; updating each node corresponding to the entity and the attribute value of each edge in the relationship graph already constructed through data mapping of the current business data, to obtain the latest relationship graph; the relationship graph representing the type and attribute value of each entity in the business data and the relationship between each entity; based on the calculation logic configuration of the obtained index task, for each node of each level of the latest relationship graph, calculating the relationship index corresponding to each level layer by layer to obtain the target relationship index value corresponding to the index task; according to the target relationship index value, the interception of the business data of illegal entities can improve the calculation efficiency of the relationship index, thereby improving the accuracy and timeliness of information interception.

Description

Technical Field

[0001] This application relates to the field of computer technology, and to, but is not limited to, methods, apparatus, devices, and readable storage media for information interception based on graph computing. Background Technology

[0002] In information interception scenarios, relevant personnel configure various relationship metrics based on business experience. Currently, the configuration of relationship metrics is typically based on traditional technologies and platforms. For example, users can use technologies such as Hadoop or Spark, with data development technicians writing relevant logic scripts to generate relationship metric calculation tasks based on the logical requirements of the relationship metrics, thereby obtaining the relationship metric results for users to analyze and verify.

[0003] Because the logical script for relational metrics involves multiple multi-table joins, it can produce results when the data volume is small and the hierarchy is shallow. However, when dealing with large data volumes, the computational performance cannot meet the needs of the actual business update frequency. This results in low computational efficiency for relational metrics, which in turn affects the accuracy and timeliness of information interception. Summary of the Invention

[0004] The information interception method, apparatus, device, and readable storage medium based on graph computing provided in this application can improve the computational efficiency of relation indicators, thereby improving the accuracy and timeliness of information interception.

[0005] The technical solution of this application embodiment is implemented as follows:

[0006] According to one aspect of the embodiments of this application, a graph computing-based information interception method is provided, comprising:

[0007] Obtain current business data, which is the type and attribute values ​​of business-related entities currently being applied in the application scenario;

[0008] The current business data is mapped to update the attribute values ​​of each node corresponding to the entity and each edge in the constructed relationship graph to obtain the latest relationship graph; the relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationships between the entities.

[0009] Based on the acquired indicator task, for each node at each level of the latest relationship graph, the relationship indicator corresponding to each level is calculated layer by layer to obtain the target relationship indicator value corresponding to the indicator task.

[0010] Based on the target relationship index value, the business data of illegal entities can be intercepted.

[0011] According to one aspect of the embodiments of this application, a graph computing-based information interception device is provided, comprising: a data acquisition module, a graph update module, an indicator calculation module, and an information interception module, wherein...

[0012] The data acquisition module is used to acquire current business data, which is the type and attribute values ​​of business-related entities currently being applied in the application scenario.

[0013] The graph update module is used to update the attribute values ​​of each node corresponding to the entity and each edge in the constructed relationship graph by mapping the current business data, so as to obtain the latest relationship graph; the relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationship between the entities.

[0014] The indicator calculation module is used to calculate the relationship indicator corresponding to each level for each node of each level of the latest relationship graph based on the acquired indicator task, so as to obtain the target relationship indicator value corresponding to the indicator task.

[0015] The information interception module is used to intercept business data of illegal entities based on the target relationship index value.

[0016] According to one aspect of the present application, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor executes the program to implement the method described in the embodiments of the present application.

[0017] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the methods provided in the embodiments of this application.

[0018] This application provides a graph-based information interception method. First, current business data is acquired, which consists of the types and attribute values ​​of business-related entities currently being used in the application scenario. Then, the current business data is mapped to update the attribute values ​​of each node and edge in the constructed relationship graph corresponding to the entities, resulting in a new relationship graph. This graph represents the types and attribute values ​​of each entity in the business data, as well as the relationships between them. Subsequently, based on the acquired indicator task, for each node at each level of the new relationship graph, the corresponding relationship indicator is calculated layer by layer to obtain the target relationship indicator value corresponding to the indicator task. Finally, based on the target relationship indicator value, the business data of illegal entities is intercepted. This application uses a relationship graph to reflect the types and attribute values ​​of each entity in the current business data, as well as the relationships between them, and calculates the target relationship indicator value for the current business data based on this graph. By constructing a relationship graph of business data, the computing power requirements of large-scale business data can be met. Thus, when the amount of business data is large, the calculation efficiency of relationship indicators can be improved, thereby improving the accuracy and timeliness of intercepting business data of illegal entities.

[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0020] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the specification, serve to explain the technical solutions of this application. Obviously, the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0021] The flowcharts shown in the accompanying drawings are merely illustrative and do not necessarily include all content and operations / steps, nor do they necessarily have to be performed in the described order. For example, some operations / steps can be broken down, while others can be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0022] Figure 1 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 1 ;

[0023] Figure 2 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 2 ;

[0024] Figure 3 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 3 ;

[0025] Figure 4 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 1 ;

[0026] Figure 5 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 4 ;

[0027] Figure 6 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 2 ;

[0028] Figure 7 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 5 ;

[0029] Figure 8 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 3 ;

[0030] Figure 9 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 4 ;

[0031] Figure 10 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 6 ;

[0032] Figure 11 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 5 ;

[0033] Figure 12 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 6 ;

[0034] Figure 13 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 7 ;

[0035] Figure 14 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 8 ;

[0036] Figure 15A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 9 ;

[0037] Figure 16 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 ;

[0038] Figure 17 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 one;

[0039] Figure 18 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 two;

[0040] Figure 19 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 three;

[0041] Figure 20 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 Four;

[0042] Figure 21 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 five;

[0043] Figure 22 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 six;

[0044] Figure 23 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 seven;

[0045] Figure 24 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 eight;

[0046] Figure 25 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 10 Nine;

[0047] Figure 26 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 7 ;

[0048] Figure 27A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 2 ten;

[0049] Figure 28 A visual interface diagram of a relationship index configuration platform provided in this application embodiment. Figure 2 eleven;

[0050] Figure 29 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 8 ;

[0051] Figure 30 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 9 ;

[0052] Figure 31 A schematic diagram of a relationship graph provided in an embodiment of this application;

[0053] Figure 32 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 10 ;

[0054] Figure 33 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 10 one;

[0055] Figure 34 A flowchart illustrating an information interception method based on graph computing provided in this application embodiment. Figure 10 two;

[0056] Figure 35 A schematic diagram of a relation index calculation device provided in this application embodiment;

[0057] Figure 36 This application provides a schematic diagram of the structure of an information interception device based on graph computing, as shown in the embodiments of the present application.

[0058] Figure 37 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0059] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0061] In the following description, references to "some embodiments," "this embodiment," "this application embodiment," and examples, etc., describe a subset of all possible embodiments. However, it is understood that "some embodiments" may be the same subset or different subset of all possible embodiments and may be combined with each other without conflict.

[0062] If the application documents contain similar descriptions such as "first / second", the following explanation shall be added: In the following description, the terms "first / second / third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first / second / third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0063] The network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. Those skilled in the art will understand that, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0064] In information interception scenarios, business personnel will conduct multiple online verifications based on experience and configure various strategy indicators for target interception during the information interception process. Traditional information interception strategies generate relevant indicators based on the characteristics of individual individuals, and then form an information interception strategy based on single or combined indicators. However, for scenarios exhibiting clustering, such as when target individuals may use the same device or the same phone to perform multiple malicious operations, it is difficult to intercept based on community associations using the above-mentioned information interception strategies.

[0065] Individuals within a community can be described through relationship networks, such as business relationships between companies, or relationships established through email, bank cards, identification documents, or telephone calls. Compared to traditional methods, relationship metrics derived from these networks offer enhanced feature representation of each individual within the community. Combined with traditional information interception strategies, this results in more accurate predictions and better interception performance. Currently, the configuration of relationship metrics is typically based on traditional technologies and platforms. For example, users can leverage technologies like Hadoop or Spark. Data development engineers can write relevant logic scripts based on the computational logic requirements of the relationship metrics to generate calculation tasks, thus obtaining the results for analysis and verification by relevant business personnel. An example of a process for calculating relationship metrics using Hadoop or Spark is as follows:

[0066] (1) Business personnel provide relationship indicator logic based on historical experience. For example, the relationship indicator logic is to calculate the total number of users associated with all users in the three-layer range through ordering devices, registering mobile phone numbers, and registering ID card numbers.

[0067] (2) Data developers prepare the necessary upstream business data tables based on the above relational indicator logic, such as user dimension tables and fact tables for user orders and registrations. Data developers write corresponding Hive or Spark scripts based on the calculation logic of relational indicators. If calculating three-level relationships, at least three joins between the same table or multiple tables are required. At the same time, depending on the relational indicators, the script performs algorithmic calculations, such as summation and percentage calculations, and writes the relational indicator results into a result table.

[0068] (3) Business personnel will analyze the IV (information value) of the relationship indicator results in the test data. If the IV value does not meet the business requirements, the business personnel will modify the indicator requirements. The process will repeat the above process until a suitable indicator is selected for use in the strategy, which will be used for information interception in the online strategy.

[0069] When determining relationship metrics using traditional technologies and platforms, two problems arise. First, data developers often need to perform multiple verifications and modifications before metrics go live in production scenarios. Traditional technologies rely on script development capabilities, making it difficult for ordinary business personnel to flexibly configure metrics, resulting in long and inefficient adjustment cycles. Second, determining relationship metrics often requires a large amount of historical data, such as login data and order data for all users over the past two years. Thus, the data volume required for relationship metric calculation often reaches tens of billions or more. Using traditional technologies and platforms to complete relationship metric calculations requires writing logic scripts involving multiple multi-table joins. While this may yield results with smaller data volumes and shallower hierarchies, the computational performance cannot meet the needs of frequent business updates when dealing with large data volumes. This leads to low efficiency in relationship metric calculation, consequently affecting the accuracy and timeliness of information interception.

[0070] Based on this, embodiments of this application provide an information interception method based on graph computing, such as... Figure 1 As shown, the method may include the following steps 101 to 104:

[0071] Step 101: Obtain current business data. Current business data consists of the types and attribute values ​​of business-related entities currently being used in the application scenario.

[0072] In this embodiment, the current business data refers to the types and attribute values ​​of business-related entities currently being used in the application scenario. For example, this could be based on user dimension tables, user order information tables, and user registration information tables, etc. Business data can be stored in a database, an Excel spreadsheet, or a CSV file (Comma-Separated Values). This embodiment does not impose any restrictions on the storage method of business data; the specific method can be selected according to the business needs of the actual application scenario.

[0073] In this application embodiment, the current business data includes the types, attributes, and attribute values ​​of multiple entities related to the business of the application scenario.

[0074] For example, to calculate the relationship indicator "Calculate the PIN of users who applied for registration yesterday and used their IMEI", the user dimension table can be used as the business data for the aforementioned relationship indicator. For the user dimension table, the entity type can be user PIN (Personal Identification Number) and IMEI (International Mobile Equipment Identity). PIN represents a user's username in the enterprise app, and the IMEI number is used to identify each individual mobile communication device, such as a mobile phone, in a mobile phone network, essentially serving as the identity information of the mobile communication device. The entity type PIN can have multiple attributes, such as "whether it was applied for yesterday" and "PIN value". The attribute "whether it was applied for yesterday" can have different attribute values, such as "applied for yesterday" and "not applied for yesterday", etc. The attribute "PIN value" can have multiple attribute values, such as "1234", "23xi", "haha", etc. The entity type IMEI can have the attribute "IMEI code", and the attribute "IMEI code" can have multiple attribute values, such as "789775", "1546769", etc.

[0075] Step 102: Through data mapping, update the attribute values ​​of each node and edge in the constructed relationship graph corresponding to the entity, and obtain the latest relationship graph. The relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationships between the entities.

[0076] In this embodiment of the application, users can construct a relationship graph through the visual interface of the relationship indicator configuration platform. The relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationships between the entities.

[0077] For example, a relationship graph can represent the relationship between different attribute values ​​of an entity's PIN and different attribute values ​​of an entity's IMEI.

[0078] In this embodiment, data mapping is used to update the attribute values ​​of each node and edge corresponding to the entity in the constructed relationship graph using the current business data, resulting in the latest relationship graph. In other words, data mapping updates the attribute values ​​of the entities and the relationships between them in the constructed relationship graph.

[0079] Step 103: Based on the computational logic configuration of the acquired indicator task, calculate the corresponding relationship indicator for each node at each level of the latest relationship graph, and obtain the target relationship indicator value corresponding to the indicator task.

[0080] In this embodiment, users can configure the calculation logic of indicator tasks through the visual interface of the relationship indicator configuration platform. It is understood that when users configure the calculation logic of indicator tasks on the front end of the relationship indicator configuration platform, the back end of the platform will obtain the configured calculation logic.

[0081] In this embodiment of the application, the relationship graph has multiple levels. For each node of each level of the latest relationship graph, the relationship index corresponding to each level is calculated layer by layer to obtain multiple relationship index values ​​corresponding to each level in the relationship graph. Based on the multiple relationship index values, the target relationship index value corresponding to the index task is obtained.

[0082] Step 104: Based on the target relationship index value, intercept the business data of illegal entities.

[0083] In this embodiment, the target relationship indicator value is the data of an illegal entity in the current business data. For example, regarding the relationship indicator "Calculate the user PIN associated with a violating company's user PIN," the target relationship indicator value can be understood as the user PIN associated with the violating company. Relevant personnel can locate the user PIN associated with the violating company by calculating the relationship indicator, and thus obtain relevant information about the user PIN. In transaction scenarios or other scenarios, relevant personnel can intercept business data related to the user PIN associated with the violating company, such as transaction data or other data.

[0084] This application provides a graph-based information interception method. First, the acquired current business data is mapped to update the attribute values ​​of each node and edge in the constructed relationship graph corresponding to the entity, resulting in the latest relationship graph. Then, based on the acquired indicator task, for each node at each level of the latest relationship graph, the relationship indicator corresponding to each level is calculated layer by layer to obtain the target relationship indicator value corresponding to the indicator task. Finally, based on the target relationship indicator value, the business data of illegal entities is intercepted.

[0085] On the one hand, this application calculates target relationship indicator values ​​for current business data based on relationship graphs. By constructing relationship graphs for business data, the computational power requirements of large-scale business data can be met. Thus, when the amount of business data is large, the calculation efficiency of relationship indicators can be improved, thereby improving the accuracy and timeliness of intercepting business data of illegal entities.

[0086] On the other hand, users can build relationship graphs and configure relevant information for indicator tasks through the visual interface of the relationship indicator configuration platform. By providing a simple and easy-to-use relationship indicator configuration platform, the user threshold can be lowered, thus avoiding the tediousness of developers writing logic scripts. In this way, the calculation cycle of relationship indicators is effectively shortened, the calculation efficiency of relationship indicators is improved, and the timeliness of intercepting business data of illegal entities is enhanced.

[0087] This application provides an information interception method based on graph computing, such as... Figure 2 As shown, the method further includes steps 201 to 204:

[0088] Step 201: In response to the graph design instructions, define the attributes of each node and the attributes of the edges associated with each node to obtain the point-edge data structure table; each node represents an entity.

[0089] In this embodiment of the application, users can define the attributes of each node in the relation graph and the attributes of the edges associated with each node in the graph design interface on the front end of the relation index configuration platform. The back end of the relation index configuration platform obtains a point-edge data structure table based on the relation graph designed by the user in the graph design interface.

[0090] In the embodiments of this application, each node represents an entity, each edge represents the relationship between two related entities, and the node-edge data table is used to represent the attributes of each node in the relationship graph and the attributes of the edges associated with each node.

[0091] Step 202: In response to the data mapping instruction, based on the point-edge data structure table, perform data mapping on multiple business data sources to obtain the point-edge mapping configuration table.

[0092] In this embodiment of the application, users can map multiple business data sources on the data mapping interface at the front end of the relationship indicator configuration platform. The back end of the relationship indicator configuration platform obtains the point-edge mapping configuration table based on the user's data mapping instructions.

[0093] In this embodiment, the business data source can be stored in the form of a database, an Excel spreadsheet, or a CSV file. This embodiment does not impose any restrictions on the storage method of the business data source.

[0094] In this embodiment of the application, the point-edge mapping configuration represents the attribute values ​​of each node and each edge formed after multiple business data sources are mapped to the corresponding nodes and edges in the relationship graph.

[0095] Step 203: In response to the relational indicator configuration instruction, determine the calculation logic configuration of the indicator task.

[0096] In this embodiment of the application, users can configure the calculation logic of indicator tasks in the task setting interface of the front end of the relationship indicator configuration platform. The back end of the relationship indicator configuration platform obtains the calculation logic configuration of the indicator tasks based on the user's relationship indicator configuration instructions.

[0097] In the embodiments of this application, the computational logic configuration represents the rule configuration information of each node, the rule configuration information of each edge, and the configuration information of the relationship index algorithm in the relationship graph.

[0098] Step 204: In response to the computation task configuration command, obtain the scheduling information of the indicator task and the output information of the indicator results.

[0099] In this embodiment of the application, users can set the scheduling information of indicator tasks in the scheduling information configuration interface of the front end of the relationship indicator configuration platform, such as the running cycle, start time, end time, etc. of the indicator tasks. The back end of the relationship indicator configuration platform obtains the scheduling information of the indicator tasks based on the scheduling information configuration instructions of the users.

[0100] In this embodiment, users can configure the output information of the indicator results in the output information configuration interface of the front end of the relationship indicator configuration platform, such as the output method of the indicator results, the output format of the indicator results, the size of the output file, etc. The back end of the relationship indicator configuration platform obtains the scheduling information of the indicator results based on the user's output information configuration instructions.

[0101] In this embodiment, users can construct relationship graphs and configure relevant information for indicator tasks through the visual interface of the relationship indicator configuration platform. On one hand, by providing a user-friendly relationship indicator configuration platform, the barrier to entry for users is lowered, and the efficiency of generating relationship indicators is increased. On the other hand, users can flexibly configure relationship indicators through the visual interface of the relationship indicator configuration platform without writing logic scripts, avoiding the tediousness of developers writing logic scripts. This effectively shortens the calculation cycle of relationship indicators, improves the calculation efficiency of relationship indicators, and thus improves the timeliness of intercepting business data of illegal entities.

[0102] It should be noted that steps 201 to 204 are parallel steps, and one or more of steps 201 to 204 can be selected. In actual execution, the execution order of steps 201 to 204 is not important and can be adjusted according to the actual situation.

[0103] In the embodiments of this application, such as Figure 3 As shown, step 201 includes steps 301 to 303:

[0104] Step 301: Receive the map design instruction and present the map design interface based on the map design instruction.

[0105] In this embodiment of the application, the user issues a graph design instruction through the front end of the relationship index configuration platform. After receiving the graph design instruction, the relationship index configuration platform presents a graph design interface on the front end. For example... Figure 4 As shown, users can set the structure of the relationship graph through the graph design interface, specifically designing the structure of each node and each edge in the relationship graph.

[0106] In the embodiments of this application, such as Figure 4 The graph design interface shown can include a title section, a toolbar section, and a work area section. The title section can include a main title and a progress title. The main title indicates the topic of the current relationship indicator calculation, such as "Calculation of XXX Relationship Indicators." The progress title indicates the current step in the relationship indicator calculation task, such as "① Graph Design," "② Data Mapping," etc. Figure 4 As shown, the progress title "① Graph Design" is selected, indicating that the current work area on the interface is a relationship design graph designed by the user.

[0107] In this embodiment of the application, the toolbar in the atlas design interface may include some commonly used tool buttons, for example, such as Figure 4 The toolbar buttons shown are, from left to right, the Export button, Save button, Add button, Forward button, Undo button, Redo button, Lock button, Snapshot button, Help button, and Page Setup button.

[0108] In this embodiment, the export button allows exporting relevant data from the relationship graph; the save button saves the current settings; the add button adds node or edge structures, or adds a new work area for setting another relationship graph; the forward button jumps from the current step to the next step, for example, from the current step "① Graph Design" to the next step "② Data Mapping"; the undo button cancels the current settings; the redo button restores the canceled settings; the lock button locks the current work area, preventing any settings from being made to the relationship design graph in the current work area; the snapshot button takes a screenshot of the current graph design interface and saves it; the help button provides instructions for using the graph design interface; and the page settings button can include some general settings for the graph design interface, such as the interface size, font size, or font type. The toolbar buttons described above are just examples; developers can add other toolbar buttons to achieve the required functions.

[0109] Step 302: Display the received relation design graph on the graph design interface. The relation design graph represents the structure of each node and edge of the designed relation graph.

[0110] In this embodiment of the application, the graph design interface displays the relationship design graph designed by the user, and the relationship design graph represents the structure of each node and each edge of the relationship graph.

[0111] In this embodiment, users can design the structure of a relationship graph by defining the types and attributes of each node and edge in the graph design interface. It should be noted that nodes in a relationship graph can also be called entities; a node is an entity that participates in the calculation of relationship indicators. One node in the relationship graph corresponds to one entity, and the attributes of the edges between two related nodes can be used as the basis for calculating relationship indicators.

[0112] In the embodiments of this application, the user can base their actions on, for example... Figure 4 The diagram design interface shown illustrates the structure of each node in the relationship graph. For example, as shown... Figure 4 The relationship design diagram shown illustrates that users first define the type of the vertices (starting nodes or starting entities) in the relationship graph. The vertex type is set to User PIN (Personal Identification Number), which represents the user's username in the enterprise app. Users then add nodes (entities) associated with the PIN nodes in the graph design interface.

[0113] For example, a node PIN can be associated with a node Android, which can represent a mobile device using the Android operating system. This mobile device can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. A node PIN can be associated with a node IMEI (International Mobile Equipment Identity), which refers to the phone's serial number or "IMEI number." The IMEI number is used to identify each individual mobile communication device, such as a mobile phone, in a mobile phone network; it's like an ID card for the mobile communication device. In other words, each mobile communication device has a unique IMEI number. A node PIN can also be associated with a node Phone, which can represent the user's mobile phone number. A node PIN can also be associated with a node EID (Electronic Identity), which is a network identity issued to citizens by the "Ministry of Public Security's Citizen Network Identity Recognition System" based on cryptographic technology and carried by a smart security chip. It can remotely identify an individual's true identity online without disclosing their identity information. It can be understood that each user can obtain a unique EID code through application or registration for remote identification of their identity online. Figure 4 As shown, users add node Android, node IMEI, node Phone, and node EID associated with the node PIN in the graph design interface.

[0114] In this embodiment of the application, after the user completes the structure of each node in the relationship graph on the graph design interface, they continue to design the structure of each edge in the relationship graph based on the graph design interface. For example, as shown... Figure 4 The diagram shown in the graph design interface illustrates the relationship design. After adding nodes (node ​​Android, node IMEI, node Phone, and node EID) associated with the vertex (node ​​or entity) PIN in the graph design interface, users can continue to add the attributes of the corresponding edges between two related nodes (entities) in the graph design interface. The attributes of the edges can be used as the data basis for calculating relationship indicators.

[0115] For example, for the edge between a vertex (node ​​or entity) PIN and a node Android, the user can add the attribute "Used," representing a user PIN for an Android mobile device used during a transaction via a company app. For the edge between a vertex PIN and a node IMEI, the user can add the attribute "Used," representing a user PIN for a specific IMEI code used during a transaction via a company app. For the edge between a vertex PIN and a node Phone, the user can add the attributes "Used" and "Receiving Mobile Number," representing a user PIN for a specific mobile number used as the receiving mobile number during a transaction via a company app. For the edge between a vertex PIN and a node EID, the user can add the attributes "Used" and "Registered," representing a user PIN for a specific EID code used or registered during a transaction via a company app.

[0116] Step 303: Based on the relational design graph, determine the attributes of each node in the designed relational graph and the attributes of the edges associated with each node to obtain the point-edge data structure table.

[0117] In this embodiment of the application, based on the relationship design graph displayed on the graph design interface, the relationship index configuration platform can determine the attributes of each node in the designed relationship graph, as well as the attributes of the edges associated with each node, and obtain a point-edge data structure table.

[0118] In the embodiments of this application, such as Figure 4 The diagram design interface shown is merely an example. Users set the structure of the relationship graph through the diagram design interface. First, users can manually add and set the type of each node (entity) in the relationship graph, or they can obtain the type of each node in the relationship graph from a preset type library. The preset type library can be set according to the actual situation of the enterprise's business data. This embodiment of the application does not impose any restrictions on the setting method of each node type in the visualization interface; the specific method can be selected according to the implementation situation. Subsequently, users can manually add the attributes of the corresponding edges between the interconnected nodes (entities) in the previous step, or users can select the attributes of the corresponding edges between the interconnected nodes in the relationship graph from a preset attribute library. The preset attribute library can be set according to the actual situation of the enterprise's business data. This embodiment of the application does not impose any restrictions on the setting method of each edge attribute in the visualization interface; the specific method can be selected according to the implementation situation. Through the above steps, the setting of each node structure and each edge structure in the relationship graph is completed, resulting in a point-edge data structure table. The point-edge data structure table includes the attributes of each node in the designed relationship graph, as well as the attributes of the edges associated with each node.

[0119] In the embodiments of this application, such as Figure 5 As shown, step 202 includes steps 401 to 403:

[0120] Step 401: In response to the data mapping command, the interface jumps from the map design interface to the data mapping interface.

[0121] In this embodiment of the application, the user issues a data mapping instruction through the front end of the relationship indicator configuration platform. After receiving the data mapping instruction, the relationship indicator configuration platform presents the data mapping instruction on the front end. For example... Figure 6 As shown, users can select the data tables corresponding to each node and edge in the relationship graph from multiple data sources through this data mapping command. Subsequently, the relationship indicator configuration platform performs data mapping for each node and edge in the relationship graph by using the target data source selected by the user.

[0122] For example, such as Figure 6 As shown, in the work area of ​​the data mapping interface, the left column is "Vertex / Edge," which represents the attributes of each node (entity) and edge in the designed relationship graph. The middle column is "Data Source," which indicates multiple selectable data sources. These data sources can be files from a data warehouse, in Excel spreadsheet format, or in CSV (Comma Separated Values) format. Here, the data warehouse can be understood as a database. The right work area presents a schematic diagram of the data mapping.

[0123] Step 402: On the data mapping interface, in response to the selection operation of the source point or source edge in the point-edge data structure table, determine the target data source from multiple business data sources, and map the target data source to the target node or target edge corresponding to the source point or source edge.

[0124] In this embodiment of the application, for a point-edge data structure table, the user selects a source point or source edge on the data mapping interface to determine the target data source from multiple business data sources, and maps the target data source to a target node or target edge corresponding to the source point or source edge. Here, the target node is the node selected by the user on the data mapping interface, and the target edge is the edge selected by the user on the data mapping interface.

[0125] like Figure 6 As shown, users can select the target node PIN in this data mapping interface, and then select the target data source corresponding to the target node PIN in the data source column. After selection, the relationship indicator configuration platform performs data mapping between the target node PIN and the target data source based on the source node selection operation. Figure 6As shown in the enlarged box, this is a data mapping diagram. The target data source includes columns for PIN, Is_yesterday_apply, and Del_flag. PIN includes multiple attribute values ​​for the user's PIN. Is_yesterday_apply indicates whether the user's PIN was applied for yesterday. For example, if the user's PIN was applied for yesterday, Is_yesterday_apply is "yes"; if the user's PIN was not applied for yesterday, Is_yesterday_apply is "no". Del_flag represents a database identifier, such as Del_flag for DEL (delete), ADD (add), EDIT (copy), etc. The data mapping diagram shows how the data in the PIN and Is_yesterday_apply columns of the target data source is mapped to the node PINs in the constructed relationship graph. The backend of the relationship indicator configuration platform updates the attribute values ​​of the node PINs in the corresponding point-edge mapping configuration table of the relationship graph, thus completing the data mapping operation for the node PINs.

[0126] Step 403: Continue to receive the selection operation of the next source point or source edge to perform point-edge mapping until all nodes and edges in the point-edge data structure table have completed data mapping with multiple business data sources. The point-edge mapping configuration table is obtained after the mapping is completed. The point-edge mapping configuration table represents the nodes and edges formed after multiple business data sources are mapped to the corresponding target nodes and target edges.

[0127] In the embodiments of this application, such as Figure 6 The data mapping interface shown is merely an example. Based on this interface, users can determine the target data source from multiple data sources and map it to the target node or edge corresponding to the source point or edge. Through these steps, users complete the data mapping of all nodes and edges in the point-edge data structure table through multiple business data sources, resulting in a mapped point-edge mapping configuration table. It can be understood that the point-edge mapping configuration table includes the nodes and edges formed after mapping multiple business data sources to their corresponding target nodes and edges.

[0128] In the embodiments of this application, such as Figure 7 As shown, step 203 includes steps 501 to 503:

[0129] Step 501: In response to the relational indicator configuration command, the interface jumps from the data mapping interface to the task settings interface.

[0130] In this embodiment, the user issues a relationship indicator configuration command through the front end of the relationship indicator configuration platform. After receiving the command, the platform redirects from the data mapping interface to the task settings interface. Figure 8 As shown, users can add computational logic configurations for multiple indicator tasks to computational nodes (computational entities) through this task settings interface. Here, the computational entity (node) corresponds to the computational node in the relationship graph; it can be understood that the computational entity (node) is a vertex (starting node or starting entity) in the constructed relationship graph.

[0131] Step 502: On the task settings interface, in response to the instruction to add an indicator task, determine the indicator task information of the calculation entity, display the indicator task information, and present the calculation logic configuration control of the indicator task.

[0132] In this embodiment, the relationship indicator configuration platform responds to the indicator task addition instruction, determines the indicator task information of the computation entity, and displays the indicator task information and the indicator task's computation logic configuration control on the task settings interface. The indicator task information represents the indicator name and its corresponding English name, and the indicator task's computation logic configuration control can be a link to the indicator task's computation logic configuration.

[0133] For example, such as Figure 8 As shown, the task settings interface includes a detailed indicator list, which, from left to right, consists of: "Serial Number," "Calculation Entity," "Indicator Name," "Indicator English Name," and "Operation." The "Serial Number" column indicates the sequential number of the indicator task; "Calculation Entity" indicates the type of calculation entity for the indicator task; "Indicator Name" indicates the indicator name for the task; "Indicator English Name" indicates the corresponding English name for the indicator; and "Operation" is a configuration control for the indicator task's calculation logic. This control provides links to configure the calculation logic for the indicator task, allowing users to configure it by clicking on the corresponding control.

[0134] For example, such as Figure 8 As shown, users can add calculation logic configurations for multiple metric tasks (relationship metrics) to the computational entity PIN in the relationship graph. Users can add relationship metrics to the computational entity PIN such as "the number of newly applied PINs registered and used with the desired IMEI in the last 30 days", "the number of newly applied PINs registered and used with the same Android in the last 30 days", "the number of users who applied in the last 30 days associated with the EID of newly applied PINs", etc.

[0135] In this embodiment, on the task settings interface, the relationship indicator configuration platform responds to the basic information settings instruction and jumps from the task settings interface to the basic information settings interface. Based on the basic information settings interface, it obtains the indicator name of the target indicator task and the corresponding English name of the indicator. For example, a user can... Figure 8 In the task settings interface shown, double-clicking the indicator name or the English name of the indicator task will cause the indicator configuration platform to respond to the basic information settings command and redirect from the task settings interface to the basic information settings interface. For example... Figure 9 As shown, users can set the indicator name and corresponding English name of the indicator task in the basic information settings interface. For example, the indicator name is "Related Indicator 1", and its corresponding English name is "test1". It is important to note that the indicator name and English name must be unique for each relational indicator (indicator task) calculation logic configuration. Furthermore, the English name of the indicator will be used as an attribute column field in the indicator result output file. Therefore, users should choose an easily recognizable name when setting the English name of the indicator to facilitate analysis of the indicator results in the indicator output file.

[0136] Step 503: In response to the triggering of the calculation logic configuration control, determine the calculation logic configuration of the indicator task.

[0137] In this embodiment of the application, in the task configuration interface, the user can configure the calculation logic of the indicator task by clicking the calculation logic configuration control. The backend of the relation indicator configuration platform responds to the triggering of the calculation logic configuration control and determines the calculation logic configuration of the indicator task.

[0138] In one embodiment, such as Figure 10 As shown, step 503 includes steps 601 to 604:

[0139] Step 601: In response to the triggering of the calculation logic configuration control, the task settings interface is switched to the relationship indicator configuration interface.

[0140] In this embodiment, the user configures the calculation logic of the indicator task by clicking the calculation logic configuration control in the task settings interface. The relationship indicator configuration platform responds to the triggering of the calculation logic configuration control by switching from the task settings interface to the relationship indicator configuration interface. For example... Figure 11 As shown, the relationship indicator configuration interface consists of three parts, from top to bottom: "Calculation Entity Settings", "Passing Vertex / Edge Settings", and "Target Entity and Indicator Algorithm Configuration".

[0141] Step 602: On the relational indicator configuration interface, receive the rule configuration information of the calculation entity corresponding to the indicator task.

[0142] In this embodiment of the application, users can set the rule configuration information of the computing entity in the "Computational Entity Settings" section on the relational indicator configuration interface, and the backend of the relational indicator configuration platform receives the rule configuration information of the computing entity corresponding to the indicator task.

[0143] In this embodiment of the application, the rule configuration information of the computation entity includes the type of the computation entity, the type of the filtering condition of the computation entity, and the first filtering condition corresponding to the attribute of the computation entity.

[0144] In this embodiment of the application, on the relation index configuration interface, the relation index configuration platform receives the type of the computational entity and the filtering condition type of the computational entity; based on the filtering condition type of the computational entity, the relation index configuration platform responds to the computational entity filtering condition setting instruction and jumps from the relation index configuration interface to display the computational entity filtering condition setting page; on the computational entity filtering condition setting page, the relation index configuration platform receives the attributes of the computational entity and the first filtering condition corresponding to the attributes of the computational entity.

[0145] For example, based on such Figure 11 The relationship indicator configuration interface shown is where the relationship indicator configuration platform responds to the calculation entity confirmation command and receives the type of the calculation entity. As one possible approach, users can click the "Calculation Entity Type" control on the relationship indicator configuration interface. Triggered by this control, a drop-down menu appears, containing multiple candidate calculation entity types. The calculation entities are the starting nodes (starting entities) in the constructed relationship graph. It can be understood that the drop-down menu of the "Calculation Entity Type" control includes the types of the starting nodes in the constructed relationship graph.

[0146] For example, based on such Figure 11 The relationship metric configuration interface shown indicates that the relationship metric configuration platform responds to the confirmation command for the filter condition type of the calculated entity and receives the filter condition type of the calculated entity. As one possible approach, users can click the "Filter Condition Type" control on the relationship metric configuration interface, such as... Figure 12 As shown, triggered by the "Filter Condition Type" control, a drop-down menu appears on the control. This drop-down menu includes "Manual Input Conditions" and "Batch Upload Conditions." "Manual Input Conditions" allows users to manually set the first filter condition for the calculated entity through the relationship indicator configuration interface. "Batch Upload Conditions" allows users to select the desired first filter condition for the calculated entity from a preset condition library. This preset condition library includes pre-stored first filter conditions for multiple calculated entities related to business data, allowing users to automatically select the correct one.

[0147] For example, such as Figure 12 As shown, the user clicks the "Filter Condition Type" control, selects "Manual Input Condition" from the drop-down menu, and then clicks the "Calculate Entity Filter Conditions" button. The relationship indicator configuration platform responds to the command to calculate entity filter conditions, redirecting from the relationship indicator configuration interface to the entity filter condition settings page. Figure 13 As shown, the left work area of ​​the computed entity filtering condition settings page is the constructed relationship graph, and the right work area is the setting area for computed entity filtering conditions. On the computed entity filtering condition settings page, the relationship indicator configuration platform responds to the computed entity attribute confirmation command and receives the attributes of the computed entity. For example, as... Figure 13 As shown, on the Calculated Entity Filtering Conditions Settings page, when a user clicks the "Calculated Entity Attributes" control, a drop-down menu appears on the "Calculated Entity Attributes" control. This drop-down menu includes multiple attributes of the calculated entity, such as "PIN Value (String)" and "Whether it was applied yesterday (String)". Here, "String" indicates that the attributes "PIN Value" and "Whether it was applied yesterday" are of string type.

[0148] For example, in such Figure 13 On the calculated entity filtering condition settings page shown, based on the attributes of the calculated entity, the relationship indicator configuration platform responds to the command to add calculated entity filtering conditions by receiving the first filtering condition corresponding to the attributes of the calculated entity. For example, as... Figure 14 As shown, the user sets the attribute of the calculated entity PIN to "Applied yesterday (String)". Based on the attribute of the calculated entity PIN, the user sets the corresponding first filter condition based on the attribute "Applied yesterday (String)". When the user clicks control number 1, a drop-down menu appears on control number 1. This drop-down menu includes the filter rules corresponding to the type of calculated entity. The attributes of the calculated entity can be divided into string type (String) and numeric type (Double or int). For string type attributes, the drop-down menu includes two filter rules: "==" and "!=". For numeric type attributes, the drop-down menu includes six filter rules: "==", "!=", ">", "<", ">=", and "<=". Figure 14 As shown, for the attribute "Whether it was applied yesterday (String)" of the computed entity, users can set a "=" filter rule. Users can manually enter the content corresponding to the set filter rule in control 2. For example, users can manually enter "1" in control 2, indicating that the first filter condition for the computed entity is "the PIN of the entity (node) applied yesterday".

[0149] It should be noted that the types of filtering rules for attributes in this application embodiment are only one example. In actual applications, the classification of attribute types can be adjusted according to business needs. In addition, the types of filtering rules for string and numeric types in this application embodiment are only one example. In actual applications, the types of filtering rules can be adjusted according to business needs.

[0150] Step 603: On the relational indicator configuration interface, receive the rule configuration information of each node and edge traversed by the computational entity corresponding to the indicator task.

[0151] In this embodiment of the application, users can set the rule configuration information of each node and each edge traversed by the computing entity in the "Passing Vertex / Edge Settings" section on the relation index configuration interface. The backend of the relation index configuration platform receives the rule configuration information of each node and each edge traversed by the computing entity corresponding to the index task.

[0152] In this embodiment of the application, the rule configuration information of each node and each edge traversed by the computational entity includes the direction of each edge traversed by the computational entity, the maximum computational level of the relational design graph (relationship graph), the attributes of each node and each edge traversed by the computational entity, and the second filtering conditions corresponding to the attributes of each node and each edge traversed by the computational entity.

[0153] In this embodiment, on the relationship indicator configuration interface, the relationship indicator configuration platform responds to the traversed edge setting command and jumps from the relationship indicator configuration interface to the traversed edge setting interface. The traversed edge setting interface displays a relationship design graph. Based on the relationship design graph, it obtains the direction of the selected edge in the relationship design graph and the maximum computational level of the relationship design graph. Based on the maximum computational level, the relationship indicator configuration platform receives each node and each edge traversed by the computational entity in the relationship design graph. On the traversed edge setting interface, the relationship indicator configuration platform receives the edge filtering condition setting command. Based on the edge filtering condition setting command, it jumps from the relationship indicator configuration interface to the edge filtering condition setting interface. Based on the edge filtering condition setting interface, the relationship indicator configuration platform determines the attributes of the selected node or edge and the second filtering condition corresponding to the attributes of the selected node or edge.

[0154] For example, on the relationship indicator configuration interface, after a user clicks the "Click to Set Steps" button, the relationship indicator configuration platform responds to the step-edge setting instruction and jumps from the relationship indicator configuration interface to the step-edge setting interface. Figure 15As shown, a relationship design diagram is displayed on the edge setting interface. Based on the relationship design diagram, the direction of the selected edge in the relationship design diagram is obtained. For example, based on the relationship design diagram presented in the left-hand work area of ​​the edge setting interface, the user can click on the target edge in the relationship design diagram, which will be selected or highlighted by color. This embodiment does not impose any limitations on this; adjustments can be made according to actual applications. Based on the selected target edge, the user clicks the "Edge Direction" control. Based on the triggering of the "Edge Direction" control, a drop-down menu is displayed on the "Edge Direction" control, such as... Figure 16 As shown, the dropdown menu includes "Unlimited," "Incoming Edges," and "Outgoing Edges." "Outgoing Edges" indicates that the relationship metric configuration platform only considers edge relationships between the current computing entity and the next entity; "Incoming Edges" indicates that the relationship metric configuration platform only considers edge relationships between the next entity and the current computing entity; and "Unlimited" indicates that the relationship metric configuration platform considers all edge relationships between the current computing entity and the next entity. For example, regarding the relationship metric "Number of PINs associated with the computing node PIN and the node IMEI," as shown... Figure 16 As shown, the user selects the edge associated with the node PIN and node IMEI in the relationship design diagram, and selects "Unlimited" in the drop-down menu of "Edge Direction" to complete the edge direction setting.

[0155] For example, on the point-edge setting interface, the relationship indicator configuration platform responds to the maximum calculation level confirmation command and receives the maximum calculation level of the relationship design graph. For instance... Figure 17 As shown, when a user clicks the "Maximum Calculation Level" control, a drop-down menu appears on the control, triggering the selection. This menu includes "Globally Consistent Level" and "Custom Level." "Globally Consistent Level" is the most commonly used level setting method, indicating that there is no need to separately set the relationship indicator types for different levels in the relationship graph (relationship design diagram). Similarly, "Globally Consistent Level" also does not require separate settings for the filtering conditions of each node and edge in the relationship graph. "Custom Level" indicates that the user needs to make different settings for the relationship indicator types and filtering conditions for each level in the relationship graph.

[0156] In one embodiment, if the maximum computation level is a globally consistent level, on the point-edge setting interface, the relationship index configuration platform determines each node and each edge that the computation entity passes through in the relationship design graph based on the maximum number of levels in the obtained relationship design graph.

[0157] For example, when a user selects "Globally Consistent Level" for "Maximum Calculation Level," the user only needs to enter the maximum level value. The relationship indicator configuration platform will automatically generate a relationship graph (relationship design graph) showing all the nodes and edges traversed by the calculated entity. The user can then delete unnecessary edges from the relationship indicator calculation based on the platform's calculation results. For example, regarding the relationship indicator (indicator task) "Number of Enterprises with Three Degrees of Association with Violating Enterprises," the user first selects the edges associated with the calculated entity (enterprise) and the enterprise in the relationship design graph on the edge setting interface. Then, the user selects "Globally Consistent Level" as the level setting method and sets the maximum level value to 3. The right-hand step column on the edge setting interface will automatically display all the nodes and edges traversed at the level corresponding to the maximum level.

[0158] In one embodiment, if the maximum computation level is a custom level, on the point-edge setting interface, the relationship index configuration platform responds to the point-edge addition instruction to obtain the selected node and the edge corresponding to the selected node in the relationship design graph, or obtain the selected edge and the node corresponding to the selected edge in the relationship design graph; the relationship index configuration platform continues to receive the next point-edge addition instruction to obtain nodes or edges until all the nodes and edges traversed by the computation entity corresponding to the custom level in the relationship design graph are obtained, thus obtaining all the nodes and edges traversed by the computation entity in the relationship design graph.

[0159] For example, after a user selects "Custom Level" for "Maximum Calculation Level," the user needs to manually add the edges and nodes traversed by each level in the relationship design graph. In such cases... Figure 18 The interface for setting up points and edges, as shown, displays a candidate set and a selected set in the upper left corner of the work area. The candidate set represents nodes and edges recommended by the relationship indicator configuration platform, while the selected set displays nodes and edges already selected in the relationship design diagram. Users can first select the step to be set, such as step 1, in the step column of the right-hand work area of ​​the interface. Then, the relationship indicator configuration platform will recommend nodes and edges corresponding to step 1, such as... Figure 18As shown, users can right-click a recommended edge and select "Set as Passing Edge" in the pop-up option box. The recommended edge and its corresponding node will then automatically appear in the right-hand step column. Similarly, users can right-click a recommended point and select "Set as Passing Point" in the pop-up option box. The recommended node and its corresponding edge will then automatically appear in the right-hand step column. Users can add all the nodes and edges that the computational entity corresponding to the custom level passes through in the relationship design graph using the above method, thus obtaining the nodes and edges that the computational entity passes through in the relationship design graph.

[0160] In this embodiment of the application, upon accessing the point-edge setting interface, the relationship index configuration platform receives a point-edge filtering condition setting instruction. Based on this instruction, the platform redirects from the relationship index configuration interface to the point-edge filtering condition setting interface. For example... Figure 19 As shown, users first select a node or edge in the step bar on the right side of the point-edge setting interface. Then, users right-click the selected node or edge, and an option box pops up on the selected node or edge. This option box includes "View Attributes," "Set Filter Conditions," and "Delete Passing Points." "View Attributes" is used to view the attribute information of the selected node or edge; "Set Filter Conditions" is used to set the second filter condition for the selected node or edge; and "Delete Passing Points" is used to delete the selected node or edge. When a user clicks the "Set Filter Conditions" button, in response to the triggering of this button, the relationship indicator configuration platform receives the point-edge filter condition setting instruction and jumps from the relationship indicator configuration interface to display the point-edge filter condition setting interface. For example... Figure 20 As shown, in the point-edge filtering condition setting interface, users can set the attributes of the selected node or edge, as well as the second filtering condition corresponding to the attributes of the selected node or edge. The relationship indicator configuration platform determines the attributes of the selected node or edge, as well as the second filtering condition corresponding to the attributes of the selected node or edge, based on the content configured by the user.

[0161] Step 604: On the relational indicator configuration interface, receive the rule configuration information of the target entity corresponding to the indicator task, as well as the indicator algorithm configuration information of the indicator task; the rule configuration information of the calculation entity, the rule configuration information of each node and edge traversed by the calculation entity, the rule configuration information of the target entity, and the indicator algorithm configuration information of the indicator task constitute the calculation logic configuration of the indicator task.

[0162] In this embodiment of the application, users can set the rule configuration information of the target entity corresponding to the indicator task and the indicator algorithm configuration information of the indicator task in the "Target Entity and Indicator Algorithm Configuration" section on the relationship indicator configuration interface. The backend of the relationship indicator configuration platform receives the rule configuration information of the target entity corresponding to the indicator task and the indicator algorithm configuration information.

[0163] In this embodiment of the application, the rule configuration information and indicator algorithm configuration information of the target entity corresponding to the indicator task include the indicator calculation method of the indicator task, the type of the target entity, the indicator type of the indicator task, the indicator algorithm, the indicator target attribute, the attribute of the target entity, and the third filtering condition corresponding to the attribute of the target entity.

[0164] In this embodiment, on the relationship indicator configuration interface, the relationship indicator configuration platform responds to the target entity setting instruction and jumps from the relationship indicator configuration interface to the target entity setting interface. On the target entity setting interface, the relationship indicator configuration platform receives the indicator calculation method of the indicator task, the type of the target entity, the indicator type of the indicator task, the indicator algorithm, and the indicator target attribute. Based on the indicator target attribute, the relationship indicator configuration platform responds to the target entity filtering condition addition instruction and jumps from the target entity setting interface to the target entity filtering condition addition interface. On the target entity filtering condition addition interface, the relationship indicator configuration platform receives the third filtering condition corresponding to the indicator target attribute.

[0165] For example, on the relationship metric configuration interface, after a user clicks the "Click to Set Target Entity" button, the relationship metric configuration platform responds to the target entity setting instruction and redirects from the relationship metric configuration interface to the target entity setting interface. Figure 21 As shown, the left work area of ​​the target entity settings interface is the relationship design diagram, and the right work area is the step bar for setting the target entity. In the target entity settings interface, the relationship indicator configuration platform responds to the indicator calculation method confirmation command and receives the indicator calculation method. For example... Figure 21As shown, the indicator calculation methods include "single target entity," "summation," and "percentage." "Single target entity" indicates that the target entity is of a single type; "summation" indicates that the target entity is of multiple types; and "percentage" indicates that the result of the relationship indicator is in the form of a numerator / denominator, with the numerator distributed across different types of entities. For example, for the relationship indicator "calculate the entity PIN associated with the entity IMEI through the entity PIN," the corresponding "indicator calculation method" is "single target entity"; for the relationship indicator "calculate the number of entity PINs associated with the entity IMEI through the entity PIN and the number of entity IMEIs," the corresponding "indicator calculation method" is "summation"; and for the relationship indicator "calculate the ratio of the entity PIN associated with the entity IMEI through the entity PIN among all entity PINs," the corresponding "indicator calculation method" is "percentage."

[0166] In this embodiment, the relationship indicator configuration platform responds to the target entity confirmation instruction by receiving the type of the target entity. The target entity refers to the object configured for the calculation content and method of the calculation entity. For example, for the relationship indicator "the entity PIN associated with the entity PIN through the entity IMEI", the target entity corresponding to this relationship indicator is PIN; for the relationship indicator "the number of entity PINs associated with the entity PIN through the entity IMEI and the number of entity IMEIs", the target entities corresponding to this relationship indicator are PIN and IMEI.

[0167] In this embodiment of the application, the relationship indicator configuration platform responds to the indicator type confirmation instruction by receiving the indicator type of the indicator task. For example... Figure 21 As shown, the "indicator type" includes statistical and proportional types. For example, for the relationship indicator "calculate the number of non-compliant enterprises that are associated with the enterprise three times", the "indicator type" of this relationship indicator is statistical.

[0168] In this embodiment of the application, the relationship indicator configuration platform receives the indicator algorithm for the indicator task in response to the indicator algorithm confirmation instruction. For example, such as... Figure 22 As shown, when a user clicks the "Indicator Algorithm" control, a drop-down menu appears on the "Indicator Algorithm" control based on the trigger of the "Indicator Algorithm" control. The menu includes "Count", "SUM", "AVG", "MAX", "MIN" and "Quantities".

[0169] In this embodiment, the relationship indicator configuration platform responds to the indicator target attribute confirmation instruction by receiving the indicator target attribute of the indicator task. The indicator target attribute is the attribute of the target entity. For example, for the relationship indicator "calculate the number of non-compliant enterprises associated with an enterprise three-dimensionally", the attribute of the target entity of this relationship indicator is "whether it is non-compliant".

[0170] Based on the target attribute of the indicator, when a user clicks the "Add Filter Condition" button on the target entity settings interface, the relationship indicator configuration platform receives the instruction to add filter conditions for the target entity and redirects from the target entity settings interface to the target entity filter condition addition interface. On the target entity filter condition addition interface, the user sets the third filter condition corresponding to the indicator target attribute. For example, for the relationship indicator "Calculate the number of enterprises with three degrees of association with the enterprise that violate regulations," the third filter condition for this relationship indicator is "Is it a violation?" == "Yes".

[0171] For example, for a relationship metric that is "the number of entity PINs associated with an entity PIN through an entity IMEI", the target entity is only an entity PIN. Therefore, the "metric calculation method" corresponding to this relationship metric is "single target entity". Figure 23 As shown, the user first selects "Single Target Entity" as the "Indicator Calculation Method." Then, the user selects the entity PIN in "Target Entity." Specifically, the user clicks the "Target Entity" button, which will display a drop-down menu containing various nodes from the relationship diagram. The user can select the entity PIN from this drop-down menu, such as... Figure 23 As shown, the "Target Entity" field will display the entity PIN identifier. After selecting the target entity, the user continues to select the "Indicator Type" for the relationship indicator "Count of entity PINs associated with entity PIN through entity IMEI". Since this relationship indicator calculates the number of entity PINs, its "Indicator Type" is statistical. This means that for the relationship indicator "Proportion of entity PINs associated with entity PIN through entity IMEI among all entity PINs", its "Indicator Type" is proportional.

[0172] After selecting the indicator type, users then select the "Indicator Algorithm" for the relationship indicator. Clicking the "Indicator Algorithm" button will bring up a drop-down menu, which includes "Count," "SUM," "AVG," "MAX," "MIN," and "Quantities." Since this relationship indicator calculates the number of entity PINs, its "Indicator Algorithm" is a count type. After selecting the "Indicator Algorithm," users then select the "Target Attribute" for the relationship indicator, such as... Figure 23 As shown, after clicking the "Target Attribute of Indicator" box, a drop-down menu will pop up. This drop-down menu includes the target entity's attributes, such as "PIN value" and "whether it was applied yesterday." Since this relationship indicator calculates the number of entity PINs, the "Target Attribute of Indicator" is the PIN value. After selecting the "Target Attribute of Indicator," the user continues to select "Target Entity Filter Conditions." It should be noted that "Target Entity Filter Conditions" is optional. Users can select specific relationship indicators. For example, the relationship indicator "Number of entity PINs associated with entity PIN through entity IMEI" does not require additional filtering conditions for the target entity. Therefore, the user can leave the "Target Entity Filter Conditions" unset. At this point, the user has completed the target entity configuration for the relationship indicator "Number of entity PINs associated with entity IMEI through entity PIN."

[0173] For example, the relationship metric is "the number of entity PINs associated with an entity PIN through an entity IMEI and the number of entity IMEIs". The target entities are entity PINs and entity IMEIs; therefore, the "metric calculation method" for this relationship metric is "summation". For example... Figure 24 As shown, the user first selects "Sum" as the "Indicator Calculation Method," and then selects the entity PIN and entity IMEI in the "Target Entity" field. Figure 24 As shown, the "Target Entity" field will display the entity's PIN and IMEI. Figure 24As shown, the setup steps for both the target entity PIN and the target entity IMEI are the same, including "Indicator Type," "Indicator Algorithm," "Indicator Target Attribute," and "Target Entity Filtering Conditions." For the target entity IMEI, since the relationship indicator is "the number of entity IMEIs associated with the entity PIN through the entity IMEI," the user can set "Indicator Type" to statistical, "Indicator Algorithm" to counting, and "Indicator Target Attribute" to IMEI number; the "Target Entity Filtering Conditions" can be ignored. For the target entity PIN, since the relationship indicator is "the number of entity PINs associated with the entity PIN through the entity IMEI," the user can set "Indicator Type" to statistical, "Indicator Algorithm" to counting, and "Indicator Target Attribute" to PIN value; the "Target Entity Filtering Conditions" can be ignored. At this point, the user has completed the target entity configuration for the relationship indicator "the number of entity PINs associated with the entity PIN through the entity IMEI and the number of entity IMEIs."

[0174] For example, for a relationship metric that is "the ratio of entity PINs associated with entity IMEIs to all entity PINs," since this relationship metric calculates a ratio, the target entities are entity PINs and entity IMEIs, and the corresponding "metric calculation method" is "percentage." Figure 25 As shown, the user first selects "Percentage" as the "Indicator Calculation Method," then selects the physical IMEI in the "Indicator Numerator" and the physical PIN in the "Indicator Denominator." Figure 25 As shown, for the denominator IMEI of the indicator, users can set the "Indicator Type" to statistical, the "Indicator Algorithm" to count, and the "Indicator Target Attribute" to the IMEI number, leaving the "Target Entity Filtering Condition" unselected. For the denominator PIN of the indicator, users can set the "Indicator Type" to statistical, the "Indicator Algorithm" to count, and the "Indicator Target Attribute" to the PIN value, leaving the "Target Entity Filtering Condition" unselected. At this point, the user has completed the target entity configuration for the relationship indicator "Calculate the ratio of entity PINs associated with entity IMEI among all entity PINs".

[0175] In one embodiment, such as Figure 26 As shown, step 204 includes steps 701 to 704:

[0176] Step 701: In response to the scheduling information configuration command, the task settings interface is switched to the scheduling information configuration interface.

[0177] In this embodiment of the application, the relationship indicator configuration platform responds to the scheduling information configuration command by switching from the task settings interface to the scheduling information configuration interface. For example... Figure 27 As shown, the scheduling information configuration interface includes "Running Cycle," "Initial Data Date," "Start Execution Time," "End Execution Time," and "Warning Notification Email." "Running Cycle" refers to task scheduling information related to offline calculations performed by the relationship indicator configuration platform. For example, "Running Cycle" can be set to "Daily," meaning the platform will run indicator tasks daily; or it can be set to "Custom Time," meaning the platform will only calculate indicator tasks within a specified time period. "Start Execution Time" refers to the time when the platform begins the indicator task, such as "10:00" or "2022-1-1 10:00." "End Execution Time" refers to the time when the platform ends the indicator task, such as "22:22" or "2022-1-2 12:00." "Warning Notification Email" refers to the email sent by the platform to designated business personnel when the indicator task ends.

[0178] It should be noted that the relationship indicator configuration platform can also output the relationship indicator results to designated personnel via telephone or SMS. This application embodiment does not impose any restrictions on this, and can be adjusted accordingly based on the actual situation.

[0179] Step 702: On the scheduling information configuration interface, obtain the running cycle, first data date, start execution time, end execution time, and warning notification information of the indicator task.

[0180] Step 703: In response to the output information confirmation command, the system jumps from the scheduling information configuration interface to the output information configuration interface.

[0181] In this embodiment of the application, the relationship indicator configuration platform responds to the output information configuration command by switching from the output information interface to the output information configuration interface. For example... Figure 28 As shown, the output information configuration interface includes "Output Method," "Output Format," "Select Cluster," and a preview of the result table format. "Output Method" refers to the output method of the relationship indicator results. This output method can be a file, an image, a compressed file, a webpage, or a PDF document. This embodiment does not impose any restrictions on the output method, and specific adjustments may be made according to business needs. "Output Format" refers to the format in which the relationship indicator results are presented in the file. For example, one indicator per line, such as... Figure 28The preview of the results table format shows an example of an output result where the "Output Format" is one metric per line. "Select Cluster" refers to the number of relational indicator results to present; for example, it can be set to a large cluster (5K), meaning the relational indicator output file will contain a maximum of 5,000 relational indicator results. After completing the output configuration, users can preview multiple relational indicator results in the results table format preview window.

[0182] Step 704: On the output information configuration interface, obtain the output method, output format, and output file size of the indicator results; the running cycle, first data date, start execution time, end execution time, and warning notification information of the indicator task are the scheduling information of the indicator task, and the output method, output format, and output file size of the indicator results are the output information of the indicator results.

[0183] In one embodiment, a graph-based information interception method is provided, such as... Figure 29 As shown, the graph-based information interception method further includes steps 801 to 803:

[0184] Step 801: Determine the point file and edge file based on the point-edge mapping configuration table.

[0185] In this embodiment of the application, the user defines the attributes of each node based on the graph design interface of the relation index configuration platform, obtains the point-edge data structure table, and the attributes of the edges associated with each node. Through the data mapping interface, the user performs data mapping on multiple business data sources based on the point-edge data structure table, and obtains the point-edge mapping configuration table.

[0186] Understandably, the point file includes the type and attribute values ​​of each node in the relational graph to be constructed, while the edge file includes the information of the starting and target nodes of each edge in the relational graph, the type of each edge, and the attribute values. The information of the starting and target nodes can be abstract data of the starting and target nodes, such as numbers or IDs.

[0187] Step 802: Convert the point file, edge file, and the calculation logic configuration of the indicator task into abstract point file, abstract edge file, and abstract rule string through data mapping.

[0188] In this embodiment, the user defines the calculation logic configuration of the indicator task based on the task setting interface of the relationship indicator configuration platform.

[0189] In this embodiment of the application, after responding to the user's graph design instructions, data mapping instructions and relation indicator configuration instructions, the backend of the relation indicator configuration platform will process the received point files, edge files and indicator task calculation logic configuration through data processing, and abstract the data of point files, edge files and calculation logic configuration into abstract point files, abstract edge files and abstract rule strings.

[0190] Step 803: Construct a relation graph based on the abstract point file, abstract edge file, and abstract rule string.

[0191] In this embodiment of the application, the backend of the relationship indicator configuration platform constructs a relationship graph based on abstract point files, abstract edge files, and abstract rule strings.

[0192] In one embodiment, the abstract rule string includes: a first abstract rule string; the first abstract rule string represents the type and attributes of the starting node in the relation graph, and the type and attributes of each edge associated with the starting node; such as Figure 30 As shown, step 803 includes steps 901 to 903:

[0193] Step 901: Based on the abstract edge file, obtain the type and attributes of each edge in the abstract edge file. According to the type and attributes of each edge associated with the starting node in the first abstract rule string, filter the edges in the abstract edge file that do not participate in the construction of the relation graph to obtain the first relation graph.

[0194] In this embodiment, the relationship indicator configuration platform parses the information of the starting node and target node of each edge, the type of each edge, and its attributes from the abstract edge file. The information of the starting node and target node can be the abstract data of the node, such as the node investor (0) and the node legal person (1). After the relationship indicator configuration platform loads the abstract data of the starting node and target node of each edge in the parsed relationship graph, as well as the type and attributes of each edge, into the backend, the relationship indicator configuration platform filters out the edges that do not participate in the construction of the relationship graph according to the type and attributes of each edge associated with the starting node as specified in the first abstract rule string, and obtains the first relationship graph.

[0195] It should be noted that during the process of obtaining the first relationship graph in the backend of the relationship indicator configuration platform, the graph structure can be compressed in the manner of CSC / CSR (Compressed Sparse Columns / Compressed Sparse Row, sparse matrix storage format).

[0196] Step 902: Based on the abstract point file, load the type and attributes of each node in the abstract point file into the first relation graph to obtain the second relation graph.

[0197] In this embodiment of the application, after the backend of the relationship index configuration platform constructs the first relationship graph, it parses the type and attribute value of each node from the abstract point file, loads the parsed node type and attribute value into the corresponding node in the first relationship graph, and obtains the second relationship graph.

[0198] Step 903: Based on the type and attributes of the starting node in the second relation graph, and according to the type and attributes of the starting node of the first abstract rule string, filter the nodes in the second relation graph that do not participate in the construction of the relation graph to obtain the relation graph.

[0199] In this embodiment, after obtaining the second relationship graph, the relationship index configuration platform continues to parse the type and attributes of the starting node of the first abstract rule string, and filters out the nodes and edges in the second relationship graph that do not participate in the construction of the relationship graph, thus obtaining the final relationship graph. At this point, the construction of the relationship graph is complete.

[0200] For example, indicator task 1 is "calculate the total amount invested by an investor in enterprises within a two-dimensional range," and indicator task 2 is "calculate the total amount invested by a legal entity as a shareholder in enterprises within a two-dimensional range." The relationship indicator configuration platform parses the first abstract rule string (start_filter) from the abstract rule string. For indicator task 1, the first abstract rule string records that the type of the starting node is "investor," and for indicator task 2, the first abstract rule string records that the type of the starting node is "legal entity." Based on the abstract point file, abstract edge file, and the first abstract rule string, the relationship indicator configuration platform constructs a relationship graph, such as... Figure 31 As shown, the starting node of the constructed relationship graph is of type investor and legal person. The attribute values ​​of each node in the relationship graph have been abstracted and processed, such as investor 1 (0) and investor 2 (1).

[0201] It should be noted that relation metric 1 and relation metric 2 use the same relation graph. Therefore, users only need to build the relation graph once. When the amount of data is small, the backend of the relation metric configuration platform can perform parallel calculations on relation metric 1 and relation metric 2.

[0202] In one embodiment, the abstract rule string includes: a second rule string; the second rule string includes rule configuration information of the computational entity configured by the computational logic, rule configuration information of each node and edge traversed by the computational entity, and rule configuration information of the target entity; such as Figure 32 As shown, step 103 includes steps 1001 to 1003:

[0203] Step 1001: Based on the rule configuration information of the computational entity in the second rule string, the rule configuration information of each node and edge traversed by the computational entity, and the rule configuration information of the target entity, obtain the hierarchical filtering rules corresponding to each level in the latest relational graph.

[0204] In this embodiment, the rule configuration information for the computational entity includes: the type of the computational entity, the type of filtering condition for the computational entity, the attributes of the computational entity, and the first filtering condition corresponding to the attributes of the computational entity. The rule configuration information for each node and edge traversed by the computational entity and the rule configuration information for the target entity include: the direction of each edge traversed by the computational entity, the maximum computation level, the attributes of each node and edge in the relationship graph, and the third filtering condition corresponding to the attributes of each node and edge in the relationship graph. The rule configuration information for the target entity includes: the type of the target entity, the attributes of the target entity, and the third filtering condition corresponding to the attributes of the target entity. The hierarchical filtering rules are the filtering rules corresponding to each level in the relationship graph, and the hierarchical filtering rules include the maximum computation level, the type and attributes of the starting node and target node in each level, the direction of the edges associated with the starting node and target node, and the type and attributes of each edge.

[0205] For example, the relationship indicator configuration platform parses the second abstract rule string (calculate_filter) from the abstract rule string. For indicator task 1, "Calculate the total amount invested by investors in companies within two degrees of separation," the constructed relationship graph is as follows: Figure 31 As shown. The second abstract rule string includes: the maximum calculation level is 2; the hierarchical filtering rules corresponding to level 1 include: the direction of the edge is investor → enterprise, the type of the starting node of level 1 is investor, and the type of the target node of level 1 is enterprise; the hierarchical filtering rules corresponding to level 2 include: the direction of the edge is enterprise → enterprise, the type of the starting node of level 2 is enterprise, and the type of the target node of level 2 is enterprise. For indicator task 2 "calculate the total amount of investment of a legal person as a shareholder in enterprises within two degrees", the second abstract rule string includes: the maximum calculation level is 2; the hierarchical filtering rules corresponding to level 1 include: the direction of the edge is legal person → enterprise, the type of the starting node of level 1 is legal person, and the type of the target node of level 1 is enterprise; the hierarchical filtering rules corresponding to level 2 include: the direction of the edge is enterprise → enterprise, the type of the starting node of level 2 is enterprise, and the type of the target node of level 2 is enterprise.

[0206] Step 1002: For each node at each level of the latest relationship graph, calculate the relationship index corresponding to each level layer by layer according to the level filtering rules corresponding to each level, and obtain multiple relationship index values.

[0207] In this embodiment of the application, after mapping the current business data, the attribute values ​​of each node and each edge in the constructed relationship graph are updated to obtain the latest relationship graph.

[0208] Understandably, the latest relationship graph can also be called the updated relationship graph, relationship graph, current relationship graph, etc.

[0209] For each node at each level of the latest relationship graph, the relationship index corresponding to each level is calculated layer by layer according to the hierarchical order of the relationship graph and the corresponding hierarchical filtering rules, resulting in multiple relationship index values. It can be understood that each level in the relationship graph has a corresponding relationship index value.

[0210] Step 1003: Based on multiple relation index values, obtain the target relation index value corresponding to the index task.

[0211] In this embodiment, the abstract rule string includes a third rule string; the third rule string includes the indicator algorithm configuration information of the indicator task in the calculation logic configuration; the relationship indicator configuration platform first parses the third rule string from the abstract rule string; then, based on the indicator calculation method, indicator type, and indicator algorithm of the indicator task in the indicator algorithm configuration information, the relationship indicator configuration platform calculates multiple relationship indicator values ​​to obtain the abstract target relationship indicator value; finally, the relationship indicator configuration platform performs data transformation on the abstract target relationship indicator value to obtain the target relationship indicator value corresponding to the indicator task.

[0212] In one embodiment, such as Figure 33 As shown, step 1002 includes steps 2001 to 2004:

[0213] Step 2001: For each node at each level of the latest relationship graph, determine the current level according to the hierarchical order of the latest relationship graph.

[0214] Step 2002: Based on the current level, according to the level filtering rules corresponding to the current level in the second abstract rule string, determine the information of each node and each edge corresponding to the current level in the latest relation graph.

[0215] In this embodiment, based on the current level, the relationship index configuration platform first obtains the type and attributes of each node, the type and attributes of each edge, and the direction of each edge from the level filtering rules corresponding to the current level in the second abstract rule string. Then, based on the type and attributes of each node, the type and attributes of each edge, and the direction of each edge in the current level, it filters out the nodes and edges in the latest relationship graph that will not participate in the calculation, and obtains the information of each node and each edge in the current level. The information of each node and each edge in the current level includes the attribute values ​​of each node, the attribute values ​​of each edge, and the direction of each edge.

[0216] For example, for indicator task 1, "Calculate the total amount of investment in enterprises by investors within two degrees", for level 1, the information of each node and each edge corresponding to the current level includes: the attribute values ​​of each node, the attribute values ​​of each edge, and the direction of each edge. Among them, the attribute values ​​of each node and the direction of each edge can be represented as: Investor 1 (0) → Enterprise 1 (5), Investor 1 (0) → Enterprise 4 (8), Investor 2 (1) → Enterprise 3 (7), Investor 2 (1) → Enterprise 2 (6), Investor 2 (1) → Enterprise 5 (9), where the number in parentheses is an abstract sequence number, and the arrow indicates the direction of the edge between two nodes. Similarly, the attribute values ​​of each node and the direction of each edge in level 2 can be represented as: Enterprise 1 (5) → Enterprise 5 (9), Enterprise 4 (8) → Enterprise 3 (7), Enterprise 2 (6) → Enterprise 5 (9).

[0217] Step 2003: Based on the information of each node and each edge corresponding to the current level, calculate the relationship index of the current level and obtain the current relationship index value.

[0218] In this embodiment of the application, the backend of the relationship index configuration platform first obtains the relationship index value of the previous level. The relationship index value of the previous level is the attribute value of each edge in the previous level. Then, based on the information of each node and each edge corresponding to the current level, the relationship index value corresponding to the previous level is used as the starting node of the current level to calculate the relationship index of the current level and obtain the current relationship index value.

[0219] In this embodiment of the application, the backend of the relationship index configuration platform determines the starting node of the current level based on the information of each node and each edge corresponding to the current level; it assigns the relationship index value corresponding to the previous level and the information of the starting node of the latest relationship graph to the starting node of the current level to obtain the update starting node of the current level; based on the update starting node of the current level, it calculates the relationship index of the current level to obtain the current relationship index value.

[0220] In this embodiment of the application, the backend of the relationship index configuration platform traverses each node of the current level according to the direction of each edge in the current level based on the update start node of the current level, and obtains the attribute value of each edge in the current level; based on the information of the start node of the latest relationship graph, the attribute value of each edge in the previous level and the attribute value of each edge in the current level, the current relationship index value is obtained.

[0221] For example, for indicator task 1, "Calculate the total amount of investment in enterprises by investors within two degrees," the maximum number of levels recorded in the second abstract rule string is 2. Therefore, the relational indicator configuration platform only needs to traverse as follows: Figure 31 The diagram shows the nodes and edges in levels 1 and 2 of the relationship graph. For level 1, the level filtering rules are as follows: the direction of the edge is investor → enterprise, the type of the starting node of level 1 is investor, and the type of the target node of level 1 is enterprise. When the relationship indicator configuration platform traverses all nodes and edges of level 1, it records the attribute values ​​of the edges passed through in the target node of level 1. For example, [0, 1] is recorded on enterprise 1 (5), which means that the starting node of enterprise 1 (5) is investor 1 (0), and the attribute value of the edge between enterprise 1 (5) and investor 1 (0) is 1 (ignoring the unit of the edge attribute). Similarly, [0, 2] is recorded on enterprise 4 (8), [1, 1] is recorded on enterprise 2 (6), [1, 1] is recorded on enterprise 3 (7), and [1, 3] is recorded on enterprise 5 (9).

[0222] For level 2, the corresponding level filtering rules include: the direction of the edge is enterprise → enterprise, the type of the starting node of level 2 is enterprise, and the type of the target node of level 2 is enterprise. For example, the starting nodes in level 2 are marked as 0 (5, 8) and 1 (6, 7, 9). Among them, 0 (5, 8) means that the starting nodes of level 2 are enterprise 1 (5) and enterprise 4 (8), and the starting nodes of enterprise 1 (5) and enterprise 4 (8) in level 1 are investor 1 (0). Similarly, 1 (6, 7, 9) means that the starting nodes of level 2 are enterprise 2 (6), enterprise 3 (7) and enterprise 5 (9), and the starting nodes of enterprise 2 (6), enterprise 3 (7) and enterprise 5 (9) in level 1 are investor 2 (1). The relational indicator configuration platform traverses each node and edge in level 2 according to the level filtering rules corresponding to level 2. After traversing each node and edge in level 2, the attribute values ​​of the edges passed through level 1 and level 2 are recorded in the target node of level 2. For example, [0, 1, 3] is recorded on enterprise 1 (5), [0, 2, 2] is recorded on enterprise 4 (8), [1, 1, 1] is recorded on enterprise 2 (6), [1, 1] is recorded on enterprise 3 (7), and [1, 3] is recorded on enterprise 5 (9). Since there are no edges that meet the conditions in enterprise 3 (7) and enterprise 5 (9), the values ​​recorded in enterprise 3 (7) and enterprise 5 (9) remain unchanged.

[0223] For example, regarding indicator task 2, "Calculate the total amount of investment by a legal entity as a shareholder in enterprises within two degrees of separation," the maximum number of levels recorded in the second abstract rule string is 2. Therefore, the relational indicator configuration platform only needs to traverse as follows: Figure 31 The diagram shows the nodes and edges in levels 1 and 2 of the relationship graph. For level 1, the level filtering rules include: the direction of the edge is legal person → enterprise, the type of the starting node of level 1 is legal person, and the type of the target node of level 1 is enterprise. When the relationship index configuration platform completes the traversal of each node and edge of level 1, it records the attribute values ​​of the edges passed through in the target node of level 1. For example, [2, 1] is recorded in enterprise 5 (9), [3, 1] is recorded in enterprise 4 (8), and [4, 1] is recorded in enterprise 3 (7). For level 2, the level filtering rules include: the direction of the edge is enterprise → enterprise, the type of the starting node of level 2 is enterprise, and the type of the target node of level 2 is enterprise. The relational indicator configuration platform traverses each node and edge in level 2 according to the level filtering rules corresponding to level 2. After traversing each node and edge in level 2, the attribute values ​​of the edges passed through level 1 and level 2 will be recorded in the target node of level 2. For example, [2, 1] is recorded in enterprise 1 (9), [3, 1, 2] is recorded in enterprise 4 (8), and [4, 1] is recorded in enterprise 3 (7).

[0224] Step 2004: Continue to calculate the relationship index value of the next level until all relationship index values ​​corresponding to the preset number of levels have been calculated, resulting in multiple relationship index values.

[0225] In this embodiment of the application, the relationship indicator configuration platform continues to calculate the relationship indicator value of the next level until the relationship indicator value corresponding to the preset number of levels is calculated, thus obtaining multiple relationship indicator values. It can be understood that each level has a corresponding relationship indicator value.

[0226] In this embodiment, after obtaining multiple relationship indicator values, the indicator algorithm configuration information recorded in the third abstract rule string is used to calculate the multiple relationship indicator values ​​to obtain the target relationship indicator value corresponding to the indicator task. For example, for indicator task 1, "calculate the total amount of investment in enterprises by an investor within two dimensions," the third abstract rule string (result_filter) includes the total investment amount. For indicator task 2, "calculate the total amount of investment in enterprises by a legal person as a shareholder within two dimensions," the third abstract rule string (result_filter) includes the total investment amount.

[0227] For indicator task 1, after the relation indicator configuration platform has traversed all levels, the endpoints that can be reached are enterprise 3 (7) and enterprise 5 (9). The result set recorded for enterprise 3 (7) includes [0, 2, 2] and [1, 1], and the result set recorded for enterprise 5 (9) includes [0, 1, 3], [1, 1, 1], and [1, 3]. For investor 1 (0), the total amount of investment in enterprises within the two-degree range is 2 + 2 + 1 + 3 = 8, and for investor 2 (1), the total amount of investment in enterprises within the two-degree range is 1 + 1 + 1 + 3 = 6. At this point, all relationship indicator values ​​for Task 1 have been calculated. The investment amounts of all investors (Investor 1 and Investor 2) are added together to obtain a final result of 8 + 6 = 14. This relationship indicator result is then written to a file. For example, if calculating the investment result for a single investor, the indicator result can be written to the output file in the format "Entity Number\tIndicator Value", with output results like "0\t8\n" and "0\t6\n". Finally, based on the output configuration information set by the user, the relationship indicator configuration platform outputs the output file to the designated user through the specified output path.

[0228] For indicator task 2, after the relationship indicator configuration platform has traversed all levels, the reachable endpoints are enterprise 5 (9), enterprise 4 (8), and enterprise 3 (7). For legal person 1 (2), the amount of investment in enterprises within the two-degree range is 1; for legal person 2 (3), the amount of investment in enterprises within the two-degree range is 1+2=3; and for legal person 3 (4), the amount of investment in enterprises within the two-degree range is 1. At this point, all indicator values ​​for indicator task 2 have been calculated. The investment amounts of all legal persons (legal person 1, legal person 2, and legal person 3) are added together to obtain the final result of 1+3+1=5. This relationship indicator result is written to a file, and the output results are "2\t1\n", "3\t3\n", and "4\t1\n". Finally, based on the output configuration information set by the user, the relationship indicator configuration platform outputs the output file to the specified user through the specified output path.

[0229] In one embodiment, such as Figure 34 As shown, step 104 includes steps 3001 to 3003:

[0230] Step 3001: Determine the output configuration file based on the scheduling information of the indicator task and the output information of the indicator results.

[0231] In this embodiment, the relationship indicator configuration platform determines the output configuration file based on the scheduling information of the indicator tasks and the output information of the indicator results. The scheduling information of the indicator tasks includes the running cycle of the indicator tasks, the date of the first data collection, the start time of execution, the end time of execution, and the warning notification information. The output information of the indicator results includes the output method of the indicator results, the output format of the indicator results, and the output file size.

[0232] Step 3002: Convert the output configuration file into an abstract output configuration file through data mapping.

[0233] In this embodiment of the application, the relational index configuration platform converts the output configuration file into an abstract output configuration file through data mapping.

[0234] Step 3003: Based on the output format and output method in the abstract output configuration file, write the target relationship index value to the specified output file according to the output format, and issue early warning information according to the output method. Based on the early warning information, the business data of illegal entities is intercepted.

[0235] In this embodiment, the relationship indicator configuration platform, based on the output format and output method in the abstract output configuration file, writes the target relationship indicator value to a specified output file according to the output format, and issues a warning message according to the output method. Based on the warning message, it intercepts the business data of illegal entities. It should be noted that the relationship indicator configuration platform can also output the relationship indicator results to designated personnel via telephone or SMS. This embodiment does not impose any restrictions on this, and adjustments can be made according to actual circumstances.

[0236] In one embodiment, a relation index calculation device (graph-based information interception device) is provided to implement a graph-based information interception method, such as... Figure 35 The diagram shows the system principle of the relationship indicator calculation device. The device includes a relationship indicator configuration platform and a graph calculation engine. Users can configure relevant information for indicator tasks through the configuration platform. The graph calculation engine is used to construct a relationship graph and calculate relationship indicators based on the user's configuration information. When a user uses the relationship indicator calculation device for the first time, they must first construct a relationship graph based on business data and then calculate relationship indicators based on this constructed graph, including the following steps:

[0237] Step 1: The relation index configuration platform responds to the graph design command and presents the graph design interface. On the graph design interface, users can define the attributes of each node and the attributes of the edges associated with each node, obtaining a point-edge data structure table. The defined information of each node and each edge is stored in the form of the point-edge data structure table.

[0238] Step 2: The relationship indicator configuration platform responds to the data mapping command and jumps from the graph design interface to the data mapping interface. Users map multiple current business data sources to each node and edge in the relationship graph corresponding to the point-edge data structure table, thus obtaining the point-edge mapping configuration table.

[0239] Step 3: In response to the relation indicator configuration command, the relation indicator configuration platform jumps from the data mapping interface to the task setting interface, where users configure the calculation logic for multiple indicator tasks.

[0240] Step 4: In response to the task configuration command, the relationship indicator configuration platform switches from the task settings interface to the scheduling information configuration interface. Users set the scheduling information for the indicator task on the scheduling information configuration interface. In response to the output information confirmation command, the relationship indicator configuration platform switches from the scheduling information configuration interface to the output information configuration interface. Users set the scheduling information for the indicator results on the output information configuration interface.

[0241] Step 5: After users complete the graph design, data mapping, relation index configuration, and calculation task configuration on the relation index configuration platform, the relation index configuration platform obtains the point file and edge file according to the point-edge mapping configuration table, obtains the rule string according to the calculation logic of the index task, and obtains the output configuration file according to the scheduling information of the index task and the output information of the index result.

[0242] Step Six: The relation index configuration platform converts the point files, edge files, rule strings, and output configuration file data into abstract point files, abstract edge files, abstract rule strings, and abstract output configuration files.

[0243] Step 7: When the platform executes the calculation task, it transmits the abstract point file, abstract edge file, abstract rule string, and abstract output configuration file to the graph computing engine. The graph computing engine then constructs the relationship graph and calculates the relationship indicators.

[0244] Step 8: The graph computing engine parses the abstract rule string and constructs a relation graph based on the parsed abstract rule string, abstract vertex file, and abstract edge file.

[0245] Step 9: Based on the constructed relationship graph, the graph computing engine calculates the relationship index according to the filtering rules defined in the parsed abstract rule string, and obtains the abstract target relationship index value.

[0246] Step 10: The relation indicator configuration platform reverse-processes the abstract target relation indicator values ​​obtained by the graph computing engine into business data to obtain target relation indicator values. The relation indicator configuration platform then issues early warning information based on the output rules specified in the abstract output configuration file, thereby intercepting business data of illegal entities.

[0247] When users utilize the relationship indicator calculation device, if the relationship graph has already been constructed, they only need to update the current business data source to the corresponding node-edge mapping configuration table of the constructed relationship graph through a data merging full snapshot operation. This can be understood as updating the data in the node and edge files based on the current business data source. In this way, the graph calculation engine only needs to update each node and its attribute values ​​in the constructed relationship graph according to the current business data. The graph calculation engine then calculates the relationship indicators based on the updated relationship graph.

[0248] This application provides a relational indicator calculation device. On one hand, it uses a relational graph to reflect the types and attribute values ​​of various entities in the current business data, as well as the relationships between these entities. Based on this relational graph, it calculates target relational indicator values ​​for the current business data. By constructing a relational graph for the business data, the computational power requirements of large-scale business data can be met. Thus, when the amount of business data is large, the calculation efficiency of relational indicators can be improved, thereby enhancing the accuracy and timeliness of intercepting business data from illegal entities. On the other hand, this application provides a relational indicator configuration platform to obtain indicator tasks. Users can construct relational graphs and configure relevant information for indicator tasks through the platform's visual interface. By providing a simple and easy-to-use relational indicator configuration platform, the user threshold can be lowered, avoiding the tediousness of developers writing logic scripts. This effectively shortens the calculation cycle of relational indicators, improves their calculation efficiency, and thus enhances the timeliness of intercepting business data from illegal entities.

[0249] In this application embodiment, a graph computing-based information interception device 5000 is provided, such as... Figure 36 As shown, the device includes a data acquisition module 5001, a map update module 5002, an indicator calculation module 5003, an information interception module 5004, a task acquisition module 5005, and a map construction module 5006, wherein...

[0250] The data acquisition module 5001 is used to acquire current business data, which is the type and attribute values ​​of business-related entities currently being applied in the application scenario.

[0251] The graph update module 5002 is used to update the attribute values ​​of each node corresponding to the entity and each edge in the constructed relationship graph by mapping the current business data, so as to obtain the latest relationship graph; the relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationship between each entity;

[0252] The indicator calculation module 5003 is used to calculate the corresponding relationship indicators for each node at each level of the latest relationship graph based on the calculation logic configuration of the acquired indicator task, and obtain the target relationship indicator value corresponding to the indicator task.

[0253] The information interception module 5004 is used to intercept business data of illegal entities based on the target relationship index value.

[0254] In some embodiments, the task acquisition module 5005 is configured to, in response to a graph design instruction, define the attributes of each node and the attributes of the edges associated with each node to obtain a point-edge data structure table; each node represents an entity; in response to a data mapping instruction, perform data mapping on multiple business data sources based on the point-edge data structure table to obtain a point-edge mapping configuration table; in response to a relational indicator configuration instruction, determine the computational logic configuration of the indicator task; and in response to a computational task configuration instruction, acquire the scheduling information of the indicator task and the output information of the indicator results.

[0255] In some embodiments, the task acquisition module 5005 is further configured to receive a graph design instruction, present a graph design interface based on the graph design instruction, display the received relation design graph on the graph design interface, the relation design graph representing the structure of each node and each edge of the designed relation graph, and based on the relation design graph, determine the attributes of each node of the designed relation graph and the attributes of the edges associated with each node to obtain a point-edge data structure table.

[0256] In some embodiments, the task acquisition module 5005 is further configured to, in response to a data mapping instruction, jump from the graph design interface to the data mapping interface; on the data mapping interface, in response to the selection operation of the source point or source edge in the point-edge data structure table, determine the target data source from multiple business data sources, and map the target data source to the target node or target edge corresponding to the source point or source edge; continue to receive the next selection operation of the source point or source edge for point-edge mapping, until all nodes and edges in the point-edge data structure table have completed data mapping with multiple business data sources, and obtain the point-edge mapping configuration table after mapping is completed. The point-edge mapping configuration table represents the nodes and edges formed after multiple business data sources are mapped to the corresponding target nodes and target edges.

[0257] In some embodiments, the task acquisition module 5005 is further configured to, in response to a relational indicator configuration instruction, jump from the data mapping interface to the task setting interface; on the task setting interface, in response to an indicator task addition instruction, determine the indicator task information of the calculation entity, display the indicator task information, and present the calculation logic configuration control of the indicator task; in response to the triggering of the calculation logic configuration control, determine the calculation logic configuration of the indicator task.

[0258] In some embodiments, the task acquisition module 5005 is further configured to, in response to the triggering of the computation logic configuration control, jump from the task setting interface to the relationship indicator configuration interface; on the relationship indicator configuration interface, receive the rule configuration information of the computation entity corresponding to the indicator task; on the relationship indicator configuration interface, receive the rule configuration information of each node and each edge traversed by the computation entity corresponding to the indicator task; on the relationship indicator configuration interface, receive the rule configuration information of the target entity corresponding to the indicator task, and the indicator algorithm configuration information of the indicator task; the rule configuration information of the computation entity, the rule configuration information of each node and each edge traversed by the computation entity, the rule configuration information of the target entity, and the indicator algorithm configuration information of the indicator task constitute the computation logic configuration of the indicator task.

[0259] In some embodiments, the task acquisition module 5005 is further configured to receive the type of the computational entity and the filtering condition type of the computational entity on the relational index configuration interface; based on the filtering condition type of the computational entity, in response to the computational entity filtering condition setting instruction, jump from the relational index configuration interface to display the computational entity filtering condition setting page; on the computational entity filtering condition setting page, receive the attributes of the computational entity and the first filtering condition corresponding to the attributes of the computational entity, wherein the type of the computational entity, the filtering condition type of the computational entity, and the first filtering condition are the rule configuration information of the computational entity.

[0260] In some embodiments, the task acquisition module 5005 is further configured to, in response to a passing point and edge setting instruction, jump from the relationship index configuration interface to the passing point and edge setting interface on the relationship index configuration interface; display a relationship design graph on the passing point and edge setting interface; based on the relationship design graph, obtain the direction of the selected edge in the relationship design graph and the maximum computation level of the relationship design graph; based on the maximum computation level, receive each node and each edge traversed by the computation entity in the relationship design graph; receive a point and edge filtering condition setting instruction on the passing point and edge setting interface; based on the point and edge filtering condition setting instruction, jump from the relationship index configuration interface to the point and edge filtering condition setting interface; based on the point and edge filtering condition setting interface, determine the attributes of the selected node or edge and the second filtering condition corresponding to the attributes of the selected node or edge; the direction of the edge, the maximum computation level, the attributes of each node and edge in the relationship design graph, and the second filtering condition are the rule configuration information of each node and each edge traversed by the computation entity.

[0261] In some embodiments, the task acquisition module 5005 is further configured to, if the maximum computation level is a globally consistent level, determine, on the point-edge setting interface, each node and each edge traversed by the computational entity in the relational design graph based on the maximum level number of the acquired relational design graph; if the maximum computation level is a custom level, on the point-edge setting interface, in response to the point-edge addition instruction, acquire the selected node and the edge corresponding to the selected node in the relational design graph, or acquire the selected edge and the node corresponding to the selected edge in the relational design graph; continue to receive the next point-edge addition instruction to acquire nodes or edges until all nodes and each edge traversed by the computational entity corresponding to the custom level in the relational design graph are acquired, thus obtaining each node and each edge traversed by the computational entity in the relational design graph.

[0262] In some embodiments, the task acquisition module 5005 is further configured to, in response to a target entity setting instruction, jump from the relationship indicator configuration interface to the target entity setting interface on the relationship indicator configuration interface; on the target entity setting interface, receive the indicator calculation method of the indicator task, the type of the target entity, the indicator type of the indicator task, the indicator algorithm, and the indicator target attribute; based on the indicator target attribute, in response to a target entity filtering condition addition instruction, jump from the target entity setting interface to the target entity filtering condition addition interface; on the target entity filtering condition addition interface, receive the third filtering condition corresponding to the indicator target attribute; the rule configuration information and indicator algorithm configuration information of the indicator calculation method of the indicator task, the type of the target entity, the indicator type of the indicator task, the indicator algorithm, the indicator target attribute, and the third filtering condition.

[0263] In some embodiments, the task acquisition module 5005 is further configured to, in response to a scheduling information configuration instruction, jump from the task setting interface to the scheduling information configuration interface; on the scheduling information configuration interface, acquire the running cycle, first data date, start execution time, end execution time, and warning notification information of the indicator task; in response to an output information confirmation instruction, jump from the scheduling information configuration interface to the output information configuration interface; on the output information configuration interface, acquire the output method, output format, and output file size of the indicator results; the running cycle, first data date, start execution time, end execution time, and warning notification information of the indicator task constitute the scheduling information of the indicator task, and the output method, output format, and output file size of the indicator results constitute the output information of the indicator results.

[0264] In some embodiments, the graph construction module 5006 is used to determine point files and edge files based on a point-edge mapping configuration table; convert the point files, edge files, and the computational logic configuration of the indicator tasks into abstract point files, abstract edge files, and abstract rule strings through data mapping; and construct a relation graph based on the abstract point files, abstract edge files, and abstract rule strings.

[0265] In some embodiments, the graph construction module 5006 is further configured to: obtain the type and attributes of each edge in the abstract edge file based on the abstract edge file; filter the edges in the abstract edge file that do not participate in the construction of the relation graph according to the type and attributes of each edge associated with the starting node in the first abstract rule string, to obtain a first relation graph; load the type and attributes of each node in the abstract point file into the first relation graph based on the abstract point file, to obtain a second relation graph; and filter the nodes in the second relation graph that do not participate in the construction of the relation graph according to the type and attributes of the starting node in the first abstract rule string, based on the type and attributes of the starting node in the second relation graph, to obtain the relation graph.

[0266] In some embodiments, the indicator calculation module 5003 is used to obtain the hierarchical filtering rules corresponding to each level in the latest relationship graph based on the rule configuration information of the calculation entity in the second rule string, the rule configuration information of each node and each edge traversed by the calculation entity, and the rule configuration information of the target entity; for each node in each level of the latest relationship graph, according to the hierarchical filtering rules corresponding to each level, the relationship indicators corresponding to each level are calculated layer by layer to obtain multiple relationship indicator values; based on the multiple relationship indicator values, the target relationship indicator value corresponding to the indicator task is obtained.

[0267] In some embodiments, the index calculation module 5003 is further configured to: for each node of each level of the latest relationship graph, determine the current level according to the hierarchical order of the latest relationship graph; based on the current level, determine the information of each node and each edge corresponding to the current level in the latest relationship graph according to the hierarchical filtering rules corresponding to the current level in the second abstract rule string; calculate the relationship index of the current level based on the information of each node and each edge corresponding to the current level, and obtain the current relationship index value; continue to calculate the relationship index value of the next level, until the relationship index values ​​corresponding to the preset number of levels are calculated, and obtain multiple relationship index values.

[0268] In some embodiments, the index calculation module 5003 is further configured to obtain, based on the current level, the type and attributes of each node, the type and attributes of each edge, and the direction of each edge from the level filtering rules corresponding to the current level in the second abstract rule string; and based on the type and attributes of each node, the type and attributes of each edge, and the direction of each edge in the current level, filter the nodes and edges in the latest relation graph that will not participate in the calculation to obtain the information of each node and each edge in the current level, wherein the information of each node and each edge in the current level includes the attribute values ​​of each node, the attribute values ​​of each edge, and the direction of each edge.

[0269] In some embodiments, the index calculation module 5003 is further configured to obtain the relation index value of the previous level, wherein the relation index value of the previous level is the attribute value of each edge in the previous level; based on the information of each node and each edge corresponding to the current level, the relation index value corresponding to the previous level is used as the starting node of the current level to calculate the relation index of the current level and obtain the current relation index value.

[0270] In some embodiments, the index calculation module 5003 is further configured to determine the starting node of the current level based on the information of each node and each edge corresponding to the current level; assign the relationship index value corresponding to the previous level and the information of the starting node of the latest relationship graph to the starting node of the current level to obtain the update starting node of the current level; and calculate the relationship index of the current level based on the update starting node of the current level to obtain the current relationship index value.

[0271] In some embodiments, the index calculation module 5003 is further configured to, based on the update start node of the current level, traverse each node of the current level according to the direction of each edge in the current level to obtain the attribute value of each edge in the current level; and obtain the current relationship index value based on the information of the start node of the latest relationship graph, the attribute value of each edge in the previous level and the attribute value of each edge in the current level.

[0272] In some embodiments, the indicator calculation module 5003 is further configured to calculate multiple relation indicator values ​​based on the indicator calculation method, indicator type, and indicator algorithm of the indicator task in the indicator algorithm configuration information, to obtain an abstract target relation indicator value; and to perform data conversion on the abstract target relation indicator value to obtain a target relation indicator value corresponding to the indicator task.

[0273] In some embodiments, the information interception module 5004 is further configured to determine an output configuration file based on the scheduling information of the indicator task and the output information of the indicator result; convert the output configuration file into an abstract output configuration file through data mapping; write the target relationship indicator value into a specified output file according to the output format and output method in the abstract output configuration file, and issue a warning message according to the output method, thereby intercepting the business data of illegal entities based on the warning message.

[0274] The descriptions of the above device embodiments are similar to those of the above method embodiments, and have similar beneficial effects. For technical details not disclosed in the device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0275] It should be noted that the module division of the graph computing-based information interception device in this application embodiment is illustrative and only represents a logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, exist as separate physical units, or have two or more units integrated into one unit. The integrated units can be implemented in hardware, as software functional units, or a combination of software and hardware.

[0276] It should be noted that, in the embodiments of this application, if the above-described methods are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0277] This application provides an electronic device. Figure 37 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 37 As shown, the electronic device 6000 includes a memory 6001 and a processor 6002. The memory 6001 stores a computer program that can run on the processor 6002. When the processor 6002 executes the program, it implements the steps in the graph computing-based information interception method provided in the above embodiments.

[0278] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps in the graph-based information interception method provided in the above embodiments.

[0279] This application provides a computer program product containing instructions that, when run on a computer, causes the computer to execute the steps in the graph computing-based information interception method provided in the above-described method embodiments.

[0280] It should be understood that the phrases "one embodiment," "an embodiment," or "some embodiments" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to an embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment," "in one embodiment," or "in some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely for descriptive purposes and do not represent the superiority or inferiority of the embodiments. The descriptions of the various embodiments above tend to emphasize the differences between the various embodiments; their similarities or commonalities can be referred to mutually, and for the sake of brevity, they will not be repeated here.

[0281] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three kinds of relationships. For example, object A and / or object B can represent three situations: object A exists alone, object A and object B exist simultaneously, and object B exists alone.

[0282] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0283] In addition, each functional module in the various embodiments of this application can be integrated into one processing unit, or each module can be a separate unit, or two or more modules can be integrated into one unit; the integrated modules can be implemented in hardware or in the form of hardware plus software functional units.

[0284] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0285] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause an electronic device to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0286] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An information interception method based on graph computation, characterized in that, The method includes: Obtain current business data, which is the type and attribute values ​​of business-related entities currently being applied in the application scenario; The current business data is mapped to update the attribute values ​​of each node corresponding to the entity and each edge in the constructed relationship graph to obtain the latest relationship graph; the relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationships between the entities. Based on the computational logic configuration of the acquired indicator task, for each node at each level of the latest relationship graph, the relationship indicator corresponding to each level is calculated layer by layer to obtain the target relationship indicator value corresponding to the indicator task; wherein, the relationship indicator is calculated layer by layer according to the level filtering rules corresponding to each level; the target relationship indicator value is the data of illegal entities in the current business data; Based on the target relationship index value, the business data of illegal entities can be intercepted.

2. The method according to claim 1, characterized in that, The method further includes: In response to the graph design instructions, the attributes of each node and the attributes of the edges associated with each node are defined to obtain a point-edge data structure table; each node represents an entity. In response to the data mapping instruction, based on the point-edge data structure table, data mapping is performed on multiple business data sources to obtain the point-edge mapping configuration table; In response to the relational indicator configuration instruction, determine the computational logic configuration of the indicator task; In response to the computation task configuration command, the scheduling information of the indicator task and the output information of the indicator results are obtained.

3. The method according to claim 2, characterized in that, In response to the graph design instructions, the attributes of each node and the attributes of the edges associated with each node are obtained to obtain a point-edge data structure table, including: Receive a map design instruction and present a map design interface based on the map design instruction; The received relation design graph is displayed on the graph design interface. The relation design graph represents the structure of each node and each edge of the designed relation graph. Based on the relationship design graph, the attributes of each node in the relationship graph and the attributes of the edges associated with each node are determined to obtain the point-edge data structure table.

4. The method according to claim 3, characterized in that, In response to the data mapping instruction, based on the point-edge data structure table, data mapping is performed on the business data source to obtain the point-edge mapping configuration table, including: In response to the data mapping command, the interface jumps from the map design interface to the data mapping interface. On the data mapping interface, in response to the selection operation of the source point or source edge in the point-edge data structure table, the target data source is determined from multiple business data sources, and the target data source is mapped to the target node or target edge corresponding to the source point or the source edge. Continue receiving the selection operation of the next source point or source edge to perform point-edge mapping until all nodes and edges in the point-edge data structure table have completed data mapping with the multiple business data sources, and obtain the point-edge mapping configuration table after mapping is completed. The point-edge mapping configuration table represents each node and each edge formed after the multiple business data sources are mapped to the corresponding target nodes and target edges.

5. The method according to claim 4, characterized in that, The step of determining the computational logic configuration of the indicator task in response to the relational indicator configuration instruction includes: In response to the relational indicator configuration command, the data mapping interface is switched to the task settings interface. On the task settings interface, in response to the instruction to add an indicator task, the indicator task information of the calculation entity is determined, and the indicator task information is displayed, as well as the calculation logic configuration control of the indicator task is presented. In response to the triggering of the calculation logic configuration control, the calculation logic configuration of the indicator task is determined.

6. The method according to claim 5, characterized in that, The step of determining the calculation logic configuration of the indicator task in response to the triggering of the calculation logic configuration control includes: In response to the triggering of the calculation logic configuration control, the interface jumps from the task settings interface to the relationship indicator configuration interface; On the relationship indicator configuration interface, the rule configuration information of the computing entity corresponding to the indicator task is received; On the relationship indicator configuration interface, the rule configuration information of each node and each edge traversed by the computing entity corresponding to the indicator task is received. On the relational indicator configuration interface, the rule configuration information of the target entity corresponding to the indicator task and the indicator algorithm configuration information of the indicator task are received; the rule configuration information of the computing entity, the rule configuration information of each node and each edge traversed by the computing entity, the rule configuration information of the target entity and the indicator algorithm configuration information of the indicator task constitute the computing logic configuration of the indicator task.

7. The method according to claim 6, characterized in that, The step of receiving rule configuration information for the computational entity corresponding to the indicator task on the relationship indicator configuration interface includes: On the relationship index configuration interface, the type of the computational entity and the type of filtering conditions for the computational entity are received; Based on the filtering condition type of the computational entity, in response to the computational entity filtering condition setting instruction, the interface jumps from the relationship index configuration interface to the computational entity filtering condition setting page. On the computing entity filtering condition setting page, the attributes of the computing entity and the first filtering condition corresponding to the attributes of the computing entity are received. The type of the computing entity, the filtering condition type of the computing entity, and the first filtering condition are rule configuration information for the computing entity.

8. The method according to claim 6, characterized in that, The step of receiving rule configuration information for each node and edge traversed by the computational entity corresponding to the indicator task on the relationship indicator configuration interface includes: On the relationship index configuration interface, in response to the through-point edge setting command, the relationship index configuration interface will jump to the through-point edge setting interface. The relationship design graph is displayed on the point-edge setting interface. Based on the relationship design graph, the direction of the selected edge in the relationship design graph and the maximum calculation level of the relationship design graph are obtained. Based on the maximum computation level, the computational entity receives each node and each edge it passes through in the relational design graph; On the point-edge setting interface, a point-edge filtering condition setting instruction is received, and based on the point-edge filtering condition setting instruction, the interface jumps from the relationship index configuration interface to the point-edge filtering condition setting interface. Based on the point-edge filtering condition setting interface, the attributes of the selected node or edge and the second filtering condition corresponding to the attributes of the selected node or edge are determined; the direction of the edge, the maximum calculation level, the attributes of each node and edge in the relationship design graph, and the second filtering condition are the rule configuration information of each node and each edge traversed by the calculation entity.

9. The method according to claim 8, characterized in that, The step of obtaining each node and each edge traversed by the computational entity in the relational design graph based on the maximum computational level includes: If the maximum computation level is a globally consistent level, on the point-edge setting interface, based on the obtained maximum level number of the relationship design graph, determine each node and each edge that the computation entity passes through in the relationship design graph; If the maximum calculation level is a custom level, in response to the point-edge setting interface, the selected node and the edge corresponding to the selected node are obtained in the relationship design graph, or the selected edge and the node corresponding to the selected edge are obtained in the relationship design graph. Continue receiving the next point edge addition instruction to obtain nodes or edges until all nodes and edges traversed by the computational entity corresponding to the custom level in the relational design graph have been obtained, thus obtaining all nodes and edges traversed by the computational entity in the relational design graph.

10. The method according to any one of claims 6 to 9, characterized in that, The step of receiving rule configuration information for the target entity corresponding to the indicator task and indicator algorithm configuration information for the indicator task on the relationship indicator configuration interface includes: On the relationship indicator configuration interface, in response to the target entity setting command, the interface jumps from the relationship indicator configuration interface to the target entity setting interface. On the target entity setting interface, the indicator calculation method, target entity type, indicator type, indicator algorithm, and indicator target attributes of the indicator task are received. Based on the target attribute of the indicator, in response to the instruction to add target entity filtering conditions, the interface jumps from the target entity setting interface to the target entity filtering condition adding interface. On the target entity filtering condition addition interface, the system receives the third filtering condition corresponding to the indicator target attribute; the indicator calculation method of the indicator task, the type of the target entity, the indicator type of the indicator task, the indicator algorithm, the indicator target attribute, and the rule configuration information of the third filtering condition, as well as the indicator algorithm configuration information.

11. The method according to any one of claims 2 to 5, characterized in that, The step of responding to the computation task configuration instruction and obtaining the scheduling information of the indicator task and the output information of the indicator results includes: In response to the scheduling information configuration command, the task settings interface will switch to the scheduling information configuration interface. On the scheduling information configuration interface, obtain the running cycle, first data date, start execution time, end execution time, and early warning notification information of the indicator task; In response to the output information confirmation command, the interface will switch from the scheduling information configuration interface to the output information configuration interface. On the output information configuration interface, the output method, output format, and output file size of the indicator results are obtained; the running cycle, first data date, start execution time, end execution time, and early warning notification information of the indicator task are the scheduling information of the indicator task, and the output method, output format, and output file size of the indicator results are the output information of the indicator results.

12. The method according to claim 2, characterized in that, The method further includes: Based on the point-edge mapping configuration table, determine the point file and edge file; The point file, the edge file, and the computational logic configuration of the indicator task are converted into abstract point files, abstract edge files, and abstract rule strings through data mapping. The relation graph is constructed based on the abstract point file, the abstract edge file, and the abstract rule string.

13. The method according to claim 12, characterized in that, The abstract rule string includes: a first abstract rule string; the first abstract rule string represents the type and attributes of the starting node in the latest relation graph, as well as the type and attributes of each edge associated with the starting node; The construction of the relation graph based on the abstract point file, the abstract edge file, and the abstract rule string includes: Based on the abstract edge file, the type and attributes of each edge in the abstract edge file are obtained. According to the type and attributes of each edge associated with the starting node in the first abstract rule string, the edges in the abstract edge file that do not participate in the construction of the relation graph are filtered to obtain the first relation graph. Based on the abstract point file, the type and attributes of each node in the abstract point file are loaded into the first relation graph to obtain the second relation graph; Based on the type and attributes of the starting node in the second relationship graph, and according to the type and attributes of the starting node of the first abstract rule string, the nodes in the second relationship graph that do not participate in the construction of the relationship graph are filtered out to obtain the relationship graph.

14. The method according to claim 12, characterized in that, The abstract rule string includes: a second rule string; the second rule string includes the rule configuration information of the computational entity configured by the computational logic, the rule configuration information of each node and edge traversed by the computational entity, and the rule configuration information of the target entity; The computational logic configuration based on the acquired indicator task calculates the corresponding relationship indicator for each node at each level of the latest relationship graph, layer by layer, to obtain the target relationship indicator value corresponding to the indicator task, including: Based on the rule configuration information of the computational entity in the second rule string, the rule configuration information of each node and edge traversed by the computational entity, and the rule configuration information of the target entity, the hierarchical filtering rules corresponding to each level in the latest relation graph are obtained. For each node at each level of the latest relationship graph, the relationship index corresponding to each level is calculated layer by layer according to the level filtering rules corresponding to each level, and multiple relationship index values ​​are obtained. Based on the multiple relationship index values, the target relationship index value corresponding to the index task is obtained.

15. The method according to claim 14, characterized in that, For each node at each level of the latest relationship graph, the relationship index corresponding to each level is calculated layer by layer according to the level filtering rules corresponding to each level, resulting in multiple relationship index values, including: For each node at each level of the latest relationship graph, the current level is determined according to the hierarchical order of the latest relationship graph; Based on the current level, and according to the level filtering rules corresponding to the current level of the second abstract rule string, the information of each node and each edge corresponding to the current level is determined in the latest relation graph. Based on the information of each node and each edge corresponding to the current level, calculate the relationship index of the current level to obtain the current relationship index value; Continue calculating the relationship index value for the next level until all relationship index values ​​corresponding to the preset number of levels have been calculated, thus obtaining the multiple relationship index values.

16. The method according to claim 15, characterized in that, The step of determining the information of each node and each edge corresponding to the current level in the latest relation graph based on the current level and according to the level filtering rules corresponding to the current level of the second abstract rule string includes: Based on the current level, obtain the node type and attributes, edge type and attributes, and edge direction in the current level from the level filtering rules corresponding to the current level of the second abstract rule string; Based on the node types and attributes, edge types and attributes, and edge directions in the current level, the nodes and edges that are not involved in the calculation in the latest relation graph are filtered to obtain the information of each node and edge in the current level. The information of each node and edge in the current level includes the attribute values ​​of each node, the attribute values ​​of each edge, and the direction of each edge.

17. The method according to claim 15, characterized in that, The step of calculating the relationship index of the current level based on the information of each node and each edge corresponding to the current level, and obtaining the current relationship index value, includes: Obtain the relation index value of the previous level, wherein the relation index value of the previous level is the attribute value of each edge in the previous level; Based on the information of each node and each edge corresponding to the current level, the relationship index value corresponding to the previous level is used as the starting node of the current level, and the relationship index of the current level is calculated to obtain the current relationship index value.

18. The method according to claim 17, characterized in that, Based on the information of each node and each edge corresponding to the current level, the relationship index value corresponding to the previous level is used as the starting node of the current level to calculate the relationship index of the current level and obtain the current relationship index value, including: Based on the information of each node and each edge corresponding to the current level, determine the starting node of the current level; Assign the relationship index value corresponding to the previous level and the information of the starting node of the latest relationship graph to the starting node of the current level to obtain the update starting node of the current level; Based on the update start node of the current level, calculate the relationship index of the current level to obtain the current relationship index value.

19. The method according to claim 18, characterized in that, The step of calculating the relationship index of the current level based on the update start node of the current level to obtain the current relationship index value includes: Based on the update start node of the current level, traverse each node of the current level according to the direction of each edge in the current level to obtain the attribute value of each edge in the current level. Based on the information of the starting node of the latest relationship graph, the attribute values ​​of each edge in the previous level, and the attribute values ​​of each edge in the current level, the current relationship index value is obtained.

20. The method according to claim 14, characterized in that, The abstract rule string includes: a third abstract rule string; the third abstract rule string includes the indicator algorithm configuration information of the indicator task configured by the computational logic; The step of obtaining the target relation indicator value corresponding to the indicator task based on the plurality of relation indicator values ​​includes: Based on the indicator calculation method, indicator type, and indicator algorithm of the indicator task according to the indicator algorithm configuration information, the multiple relationship indicator values ​​are calculated to obtain the abstract target relationship indicator value. The abstract target relationship index value is transformed to obtain the target relationship index value corresponding to the index task.

21. The method according to claim 1 or 2, characterized in that, The step of intercepting business data of illegal entities based on the target relationship index value includes: Based on the scheduling information of the indicator tasks and the output information of the indicator results, determine the output configuration file; The output configuration file is converted into an abstract output configuration file through data mapping; Based on the output format and output method in the abstract output configuration file, the target relationship index value is written to the specified output file according to the output format, and a warning message is issued according to the output method. Based on the warning message, the business data of illegal entities is intercepted.

22. An information interception device based on graph computing, characterized in that, The device includes a data acquisition module, a map update module, an indicator calculation module, and an information interception module, wherein... The data acquisition module is used to acquire current business data, which is the type and attribute values ​​of business-related entities currently being applied in the application scenario. The graph update module is used to update the attribute values ​​of each node corresponding to the entity and each edge in the constructed relationship graph by mapping the current business data, so as to obtain the latest relationship graph; the relationship graph represents the type and attribute values ​​of each entity in the business data, as well as the relationship between the entities. The indicator calculation module is used to calculate the relationship indicator corresponding to each level for each node of each level of the latest relationship graph based on the calculation logic configuration of the acquired indicator task, and obtain the target relationship indicator value corresponding to the indicator task; wherein, the relationship indicator is calculated layer by layer according to the level filtering rules corresponding to each level; the target relationship indicator value is the data of illegal entities in the current business data; The information interception module is used to intercept business data of illegal entities based on the target relationship index value.

23. An electronic device comprising a memory and a processor, the memory storing a computer program executable on the processor, characterized in that, When the processor executes the program, it implements the method according to any one of claims 1 to 21.

24. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 21.

Images