Unlock instant, AI-driven research and patent intelligence for your innovation.

Abnormal target detection method and device for heterogeneous network, equipment and storage medium

A technology for abnormal targets and heterogeneous networks, applied in abnormal target detection methods, equipment and storage media, and device fields, can solve problems such as easy to fall into local optimal solutions, limited learning accuracy, difficult to deal with naturally, and achieve Avoid the effects of accuracy, good representation, and high approximation

Active Publication Date: 2019-06-28
SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA
View PDF3 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Matrix decomposition is by decomposing the high-dimensional Laplacian matrix of the graph into two matrix products with smaller dimensions, but it is computationally expensive and difficult to deal with dynamic network problems naturally
On the other hand, traditional methods often use convex optimization technology to optimize this non-convex problem, which is easy to fall into a local optimal solution, which greatly limits the learning accuracy.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Abnormal target detection method and device for heterogeneous network, equipment and storage medium
  • Abnormal target detection method and device for heterogeneous network, equipment and storage medium
  • Abnormal target detection method and device for heterogeneous network, equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] see figure 1 , figure 1 It is a schematic flowchart of an embodiment of the method for detecting abnormal objects in a heterogeneous network of the present invention. Such as figure 1 As shown, the abnormal target detection method includes the following steps:

[0032] S11: Obtain input information data of the heterogeneous network;

[0033] Among them, the heterogeneous network can be a social network of bribery crimes, a network of obese people, a network of people with AIDS, and other social networks.

[0034] Taking the social relationship network of bribery crime as an example, the social relationship network of bribery crime is divided into nodes and edges. Among them, nodes have two types: people or companies, and edges have: directed edges (such as bribery, bribery) and undirected edges (such as Friendship) two types.

[0035] In addition, this step also has a data storage function, which can store and call the input information data of heterogeneous network...

Embodiment 2

[0081] see Figure 5 , Figure 5 It is a schematic structural diagram of an embodiment of an abnormal target detection device for a heterogeneous network of the present invention. Such as Figure 5As shown, the abnormal target detection device includes a data acquisition module 51, a model building module 52 and a target output module 53, and the target output module 53 includes a node result output unit 531, a node information analysis unit 532 and an abnormal target production unit 533.

[0082] Wherein, data acquisition module 51 is used for obtaining the input information data of heterogeneous network; Model building module 52 is used for establishing graph neural network model according to this input information data; Target output module 53 is used for based on this input information data and this graph neural network The model outputs anomalous targets.

[0083] Among them, the node result output unit 531 is used to obtain the low-dimensional characterization vector...

Embodiment 3

[0089] The present invention provides an abnormal target detection device for a heterogeneous network, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein, the memory stores instructions executable by the at least one processor, The instruction is executed by the at least one processor, so that the at least one processor can execute the method according to the first embodiment.

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an abnormal target detection method and device for a heterogeneous network, equipment and a storage medium. The abnormal target detection method comprises the steps of obtaining input information data of the heterogeneous network; Establishing a graph neural network model according to the input information data; And outputting an abnormal target based on the input information data and the graph neural network model. According to the method, the graph neural network model is established for the input information data of the heterogeneous network, the abnormal target is output based on the input information data and the graph neural network model, the graph neural network model is adopted, and the model and the actual approximation degree are high, so that the detection result is accurate. The method can be widely applied to heterogeneous network data processing and analysis detection of big data.

Description

technical field [0001] The present invention relates to the field of big data, in particular to a heterogeneous network abnormal target detection method, device, equipment and storage medium. Background technique [0002] Networks are ubiquitous in real life, such as protein networks in organisms. Proteins have different gene regulation, transcription and metabolic interactions, which are called underlying cellular networks. Abnormal protein action can lead to diseases such as mad cow disease. Whether eaten raw or cooked, it will infect the crowd and form a common disease network. This is called the middle-level human disease network. Similar genes will form a population that is resistant to certain diseases. , such as depression and anxiety. The more sensitive sub-network, the network at the top such as social network, the link between people, family relationship, friend relationship, sexually transmitted disease transmission relationship, power sex trading network, power ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/901G06N3/04G06N3/08G06Q50/00
CPCY02D10/00
Inventor 史玉回曲良黄骏
Owner SOUTH UNIVERSITY OF SCIENCE AND TECHNOLOGY OF CHINA