An application system and method for event tracing based on graph data analysis

Abstract

The present invention relates to the field of data analysis, and in particular to an application system and method based on graph data analysis in event tracing. The system includes: a graph data storage layer: the bottom-level abstract event data repository, which generates a knowledge graph based on the event itself and the attributes between events; a graph data interface service layer: creating an interface service based on event attribute retrieval requirements, docking the graph data storage layer through an API, and establishing interaction with event standardized data; a system application interface service layer: conducting a comprehensive analysis and design of requirements in different fields, performing instantiation configuration management of things, and establishing personalized data interaction with the graph data interface service layer; a data visualization editing layer: docking the business operation display interface of the system application interface service layer through an API, including modeling creation of entity data, registration of graph events, event data editing, and link-based display of graph events. This solves the limitation problem of existing data storage services in data tracing.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to an application system and method for event tracing based on graph data analysis. Background Art

[0002] With the rapid development of information technology, society has entered the era of big data. In this era, vast amounts of data are generated constantly. This vast amount of data holds a wealth of valuable information. However, due to the diverse nature and sheer volume of data, efficiently and accurately extracting this useful information is challenging. While many information extraction tools have emerged, they can improve efficiency and save time in certain scenarios. However, the relationships between data and the interconnections between knowledge bases—the so-called knowledge graphs—actually hold greater practical or commercial value. This is particularly true in specific fields, such as the investigation and tracing of water pollution incidents in the environmental protection sector or the application of financial knowledge in information security. These knowledge graphs can assist professionals in meeting their cognitive needs. Currently, there are relatively few professional visualization application systems developed based on data graph relationships, with features such as data interoperability with existing business systems in these fields.

[0003] Existing technologies often only provide a graph data storage service, and have certain limitations in providing data traceability, such as the ability to locate the source of an event in geographic space. Summary of the Invention

[0004] The present invention provides an application system and method for event tracing based on graph data analysis, aiming to address the limitations of existing data storage services in data tracing.

[0005] The present invention provides an application system for event tracing based on graph data analysis. The system structure layer includes the following from bottom to top:

[0006] Graph data storage layer: The lowest level abstract event data repository, which generates a knowledge graph based on the events themselves and the attributes between events;

[0007] Graph data interface service layer: Creates interface services based on event attribute retrieval requirements, connects to the graph data storage layer through APIs, and establishes interaction with event standardized data;

[0008] System application interface service layer: This layer comprehensively analyzes and designs requirements in different fields, manages instantiation configuration of things, and establishes personalized data interaction with the graph data interface service layer.

[0009] Data visualization editing layer: connects to the business operation display interface of the system application interface service layer through API, including model creation of entity data, registration of graph events, event data editing, and link-based display of graph events.

[0010] As a further improvement of the present invention, in the graph data interface service layer, the creation of interface services includes the creation of basic labels, edges, nodes, and data fuzzy retrieval.

[0011] As a further improvement of the present invention, the graph data storage layer is the Nebula Graph native graph database, and the graph data interface service layer creates labels, edges, and nodes through the Create Tag statement, and creates data fuzzy retrieval through the Match keyword.

[0012] As a further improvement of the present invention, the graph data interface service layer uses Python language to implement data services, integrates network request components at the service architecture level, and adopts Docker cluster load for service deployment.

[0013] The present invention also provides an application method based on graph data analysis in event tracing, which is characterized by including the following process:

[0014] Event creation and storage process:

[0015] a1. Data visualization editing input steps: Create event instances and the relationships between objects in the event instances on the visualization operation page, create tags and tag attributes on the objects, and manage data tags;

[0016] a2. System application interface service input step: Decompose the business data on the visualization operation page, including object relationships, object tags, and tag attributes, and store them in a relational database;

[0017] a3. Graph Data Interface Service Input Step: Edit business data, create corresponding nodes, edges, and labels based on the objects, object relationships, object labels, and their attributes in the business data, and return the data to the graph data repository in a standard format.

[0018] a4. Graph data storage input step: The graph data repository generates and stores the corresponding knowledge graph based on the creation definitions of nodes, edges, and labels;

[0019] Event sourcing display process:

[0020] b1. Data visualization editing and output steps: Select business data to view as needed and display it visually;

[0021] b2. System application interface service output step: Call the standard data returned by the graph data interface service, retrieve the relationship between event links from different nodes as needed, process the data, and publish the service to the data visualization editing layer;

[0022] b3. Graph data interface service output step: Perform fuzzy retrieval on the business data in the graph data repository and return the standard business data used to generate the knowledge graph for on-demand invocation by the system application interface service;

[0023] b4. Graph data storage output step: Output the standard business data for generating the knowledge graph.

[0024] As a further improvement of the present invention, the graph data repository is a Nebula Graph database. In process a3, the graph data interface service creates connections between nodes, edges, and labels on the Nebula Graph database using the Create Tag statement. In process b3, the graph data interface service performs fuzzy search on the Nebula Graph database using the Match keyword. The graph data interface service executes the corresponding statements and returns the returned data in a standard data format.

[0025] As a further improvement of the present invention, in the process a1, data tag management includes first instantiating the event, tagging it, and then creating and saving data based on the attributes of the tag.

[0026] As a further improvement of the present invention, the system application interface service is connected with the geographic spatial information data. In the process a1, the data tag management includes selecting the relationship with the tag of the corresponding spatial data as needed.

[0027] The beneficial effects of this invention include: structured representation of knowledge, abstract storage of data, business data, and events, data analysis capabilities in corresponding fields, graphical description of events, and logical organization of temporal and spatial relationships. This system allows for more flexible application of event knowledge to corresponding production fields, yielding tangible benefits. It also integrates event-centric knowledge base research with the spatial domain, enabling the display of the geographic spatial locations of relevant basic events. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a system structure diagram of the application system of the present invention based on graph data analysis in event tracing. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0030] like Figure 1 As shown, the present invention is an application system based on graph data analysis in event tracing, characterized in that the system structure layer includes, from bottom to top:

[0031] Graph data storage layer: The lowest level abstract event data repository, which generates a knowledge graph based on the events themselves and the attributes between events;

[0032] Graph data interface service layer: Creates interface services based on event attribute retrieval requirements, connects to the graph data storage layer through APIs, and establishes interaction with event standardized data;

[0033] System application interface service layer: This layer comprehensively analyzes and designs requirements in different fields, manages instantiation configuration of things, and establishes personalized data interaction with the graph data interface service layer.

[0034] Data visualization editing layer: connects to the business operation display interface of the system application interface service layer through API, including model creation of entity data, registration of graph events, event data editing, and link-based display of graph events.

[0035] From bottom to top, the system structure layer is organized with the graph data storage layer (Nebula Graph) as the underlying abstracted data storage support, capable of clustered deployment. The direct interface layer for the storage medium is an interface service layer developed specifically for its API. This separate interface service layer is built to provide for the future implementation of an open platform. System application interface services are then built upon the graph data interface service, enabling data interaction based on the graph data interface's functionality and actual needs. The data visualization editing layer is a web-based business operating system, providing services such as modeling entity data, registering graph events, and displaying graph events in a chained manner. Other functions primarily address data editing. Data visualization editing is generally divided into data tag management and data instance creation. The business logic presented can be described as follows: a knowledge graph or event is first instantiated and labeled, and then data is created and saved based on the tag's attributes.

[0036] The technical solution of the present invention mainly includes the research and application of graph data, the design and implementation of data storage interface services, and the design and implementation of application systems under different events or knowledge fields.

[0037] For graph data research and application, Nebula Graph is used as the data storage medium. Its main features are that it is an open source, distributed, and easily scalable native graph database. According to its development manual and the corresponding application environment, it can quickly implement corresponding application deployment.

[0038] For the design and implementation of data storage interface services, combined with the characteristics of graph data storage itself, interface services are developed in a coordinated manner according to the needs of corresponding fields, including basic labeling, edge, node creation, fuzzy retrieval of data and other functions, and the corresponding functions of business logic are implemented according to standard specifications.

[0039] The concepts of labels, edges, and nodes are defined in the Nebula Graph library and can be implemented according to its specified syntax. For example, tags can be created using the "Create Tag" statement, and fuzzy searches can be performed using the "Match" keyword. The interface service implementation steps first create a Nebula connection, then execute the corresponding statements and return the data in a standard format.

[0040] Because the interface service primarily connects to Nebula Graph's API for development, Python was used for service implementation, taking into account factors such as the system's deployment platform integration. At the service architecture level, in addition to integrating components like network requests, the service deployment utilizes a Docker cluster load balancing approach to meet high availability requirements.

[0041] For application system design and implementation, we conduct overall analysis and design of requirements in different fields, abstractly implement instantiation configuration management of things, and meet the scalable integration capabilities of personalized businesses.

[0042] The application system primarily addresses business scenarios and individual needs, with scalability a key design consideration. Utilizing mainstream Java technologies, the system's functional design and development leverages the Spring integration framework. In addition to implementing its own application interfaces, the application system also addresses the needs of business partners, such as creating instance data models. Business data is stored in the relational database PostgreSQL. The front-end interface utilizes the mainstream Vue framework to enhance the user experience.

[0043] Based on the system of the present invention, a method for applying graph data analysis to event tracing is also provided, including the following process:

[0044] Event creation and storage process:

[0045] a1. Data visualization editing input steps: Create event instances and the relationships between objects in the event instances on the visualization operation page, create tags and tag attributes on the objects, and manage data tags;

[0046] a2. System application interface service input step: Decompose the business data on the visualization operation page, including object relationships, object tags, and tag attributes, and store them in a relational database;

[0047] a3. Graph Data Interface Service Input Step: Edit business data, create corresponding nodes, edges, and labels based on the objects, object relationships, object labels, and their attributes in the business data, and return the data to the graph data repository in a standard format.

[0048] a4. Graph data storage input step: The graph data repository generates and stores the corresponding knowledge graph based on the creation definitions of nodes, edges, and labels;

[0049] Event sourcing display process:

[0050] b1. Data visualization editing and output steps: Select business data to view as needed and display it visually;

[0051] b2. System application interface service output step: Call the standard data returned by the graph data interface service, retrieve the relationship between event links from different nodes as needed, process the data, and publish the service to the data visualization editing layer;

[0052] b3. Graph data interface service output step: Perform fuzzy retrieval on the business data in the graph data repository and return the standard business data used to generate the knowledge graph for on-demand invocation by the system application interface service;

[0053] b4. Graph data storage output step: Output the standard business data for generating the knowledge graph.

[0054] The graph data repository is the Nebula Graph database. In process a3, the graph data interface service uses the Create Tag statement to create connections between nodes, edges, and labels on the Nebula Graph database. In process b3, the graph data interface service uses the Match keyword to perform fuzzy search on the Nebula Graph database. The graph data interface service executes the corresponding statements and returns the returned data in a standard data format.

[0055] In process a1, data tag management includes instantiating the event, tagging it, and then creating and saving data based on the tag's attributes.

[0056] The system application interface service is connected with the geographic spatial information data. In process a1, data tag management includes the selection of relationships with tags of corresponding spatial data as needed.

[0057] This invention analyzes knowledge structure and abstractly decomposes and stores it, allowing for the retrieval of event-link relationships from different nodes. In practical production applications, graph data interfaces with business systems with flexibility, loose coupling, and configurability. It provides a fundamental description of objects in specific domains, demonstrating the relationships between links, and demonstrating the logical relationships between entities in the spatiotemporal domain and their geographic distribution.

[0058] First, the present invention has the ability to connect with geospatial information data at the architecture level. Through API interface connection, such as the link to regional layer data, data can be synchronized and stored in a timely manner. The geospatial information system reflects the needs of the owner's business layer and targets its personalized characteristics. On-demand service expansion can be performed at the system application interface service layer of the architecture. In the service function expansion step, the system application layer first calls the graph data interface layer to return standard data, then processes the data as needed, and opens the interface to a dockable API for use by other docking parties.

[0059] While carrying out label management for specific things, in addition to creating relationships between the things themselves, such as the undertaking unit of a certain node, the processing results, etc., the relationship with the labels of the corresponding spatial data is selected as needed to realize the storage of relationships between labels.

[0060] After the creation of tagged knowledge data, the occurrence of events, and their progress are recorded (via API integration or input into the visual data editing page), the system can automatically maintain connections between objects based on tag relationships. Ultimately, when searching and displaying big data, objects can be displayed on demand based on their own form, including geospatial information.

[0061] Taking the spatial location of epidemic detection point data as an example, the implementation process of this system is as follows:

[0062] (1) Agree and create the detection point layer label and label attributes through the system operation page, such as the layer service call address, operator, department, etc. When the creation is successful, the graph data interface also has the ability to edit the detection point layer.

[0063] (2) After the business personnel have published the geographic information data, the business system will store the service access address in the graph database, that is, call the API of the graph data interface service.

[0064] (3) Based on the needs of the demand side, such as focusing on a certain area and displaying data of a certain time period on the App side map, the system application layer can independently implement and publish services for this demand side, so that the App caller can implement the function.

[0065] (4) Of course, other functional applications can be applied to other dimensions based on other attributes of layer tags. Or in the near future, layer tags will be associated with other event tags. This will provide the ability to search from more angles.

[0066] This patent is based on the theory of knowledge graphs, which concretizes the application of concepts and physical entities. For example, an event itself contains a kind of information that can define a specific time and a specific place and time of occurrence, or more importantly, an event and a group of events involving which roles, that is, an application system for event graph visualization. This system achieves accurate application in the description of events, the evolution of events, and the relationship between events, as well as the temporal, causal, sequential, and inclusive relationships between events. At the same time, this application has data interoperability with spatial geographic information systems, and can accurately locate the location of events.

[0067] To achieve event tracing, we must first have visual operational capabilities for events and the ability to store associations between events and their attributes. This system chose Nebula Graph as its underlying data storage medium. Nebula Graph is a reliable, distributed, linearly scalable, and highly efficient graph database. Compared to other graph databases, NebulaGraph can handle twice as much data as other graph databases, and its data storage capacity meets the needs of current big data scenarios. It provides a graph database solution with millisecond-level query latency, ensuring high availability through scalability and data recovery.

[0068] The knowledge and event graphs in this system are implemented in network topology using the vis-network component, which integrates with popular front-end technologies. It is a component in the vis.js visualization library and supports custom shapes, styles, colors, sizes, and images. Furthermore, it smoothly displays a relatively large number of nodes and their relationships on mainstream browsers, offering significant advantages in presentation quality.

[0069] Current systems similar to this example primarily feature abstracted design for custom knowledge and data entry and editing. This example, in addition to customizing graph data and displaying it in a topological format, also enables interoperability with data from corresponding domain systems, achieving a degree of scalability at the data application interface service level.

[0070] The system of the present invention has developed corresponding functional interfaces such as editing operations for the graph database, and has made certain optimizations in data editing and batch import management operations, making it more user-friendly than other application systems. In actual deployment, the system categorizes and mounts corresponding graph knowledge or events according to levels in different business fields. Moreover, for all knowledge or specific events, you can select a node and keyword search as needed, and graphically display the associated information of the data to achieve the effect of data impact analysis. In terms of integration with the domain business system, this system has a certain degree of loose coupling, adopts a unified configuration management method, and will not affect the normal application of the system due to problems such as data synchronization of a certain node.

[0071] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, and all of these should be considered to fall within the scope of protection of the present invention.

Claims

1. An application system based on graph data analysis in event tracing, characterized by: The system structure layer includes from bottom to top: Graph data storage layer: The lowest level abstract event data repository, which generates a knowledge graph based on the events themselves and the attributes between events; Graph data interface service layer: Creates interface services based on event attribute retrieval requirements, connects to the graph data storage layer through APIs, and establishes interaction with event standardized data; System application interface service layer: This layer comprehensively analyzes and designs requirements in different fields, manages instantiation configuration of things, and establishes personalized data interaction with the graph data interface service layer. Data visualization editing layer: connects to the business operation display interface of the system application interface service layer through API, including model creation of entity data, registration of graph events, event data editing, and link-based display of graph events; In the graph data interface service layer, the creation of interface services includes the creation of basic labels, edges, nodes, and data fuzzy retrieval; The graph data storage layer is the Nebula Graph native graph database. The graph data interface service layer creates tags, edges, and nodes through the CreateTag statement and creates data fuzzy retrieval through the Match keyword.

2. The application system based on graph data analysis in event tracing according to claim 1 is characterized in that: The graph data interface service layer uses Python language to implement data services, integrates network request components at the service architecture level, and adopts Docker cluster load for service deployment.

3. An application method based on graph data analysis in event tracing, characterized in that: The following processes are included: Event creation and storage process: a1. Data visualization editing input steps: Create event instances and the relationships between objects in the event instances on the visualization operation page, create tags and tag attributes on the objects, and manage data tags; a2. System application interface service input step: Decompose the business data on the visualization operation page, including object relationships, object tags, and tag attributes, and store them in a relational database; a3. Graph Data Interface Service Input Step: Edit business data, create corresponding nodes, edges, and labels based on the objects, object relationships, object labels, and their attributes in the business data, and return the data to the graph data repository in a standard format. a4. Graph data storage input step: The graph data repository generates and stores the corresponding knowledge graph based on the creation definitions of nodes, edges, and labels; Event sourcing display process: b1. Data visualization editing and output steps: Select business data to view as needed and display it visually; b2. System application interface service output step: Call the standard data returned by the graph data interface service, retrieve the relationship between event links from different nodes as needed, process the data, and publish the service to the data visualization editing layer; b3. Graph data interface service output step: Perform fuzzy retrieval on the business data in the graph data repository and return the standard business data used to generate the knowledge graph for on-demand invocation by the system application interface service; b4. Graph data storage output step: Output the standard business data for generating the knowledge graph; The graph data repository is the Nebula Graph database. In process a3, the graph data interface service uses the Create Tag statement to create connections between nodes, edges, and labels on the Nebula Graph database. In process b3, the graph data interface service uses the Match keyword to perform fuzzy search on the Nebula Graph database. The graph data interface service executes the corresponding statements and returns the returned data in a standard data format.

4. The method for applying graph data analysis to event tracing according to claim 3 is characterized in that: In the process a1, data tag management includes first instantiating an event, tagging it, and then creating and saving data based on the attributes of the tag.

5. The method for applying graph data analysis to event tracing according to claim 3 is characterized in that: The system application interface service is connected to the geographic spatial information data. In the process a1, data tag management includes selecting a relationship with the tag of the corresponding spatial data as needed.

Images