Blockchain Graph Embedding for Abnormal Behavior Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current methods for detecting blockchain abnormal behaviors, such as phishing and money laundering, are inefficient and unstable due to the need for extensive data collection and integration, leading to increased complexity and burden on detection systems.

Innovation Solution

A method using graph embedding to detect blockchain abnormal behaviors by collecting and analyzing node and transaction features, constructing a transaction graph, and establishing an abnormal behavior recognition model through biased random walk strategies, which provides a more accurate and efficient risk assessment.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If multiple detection methods are integrated to detect all possible abnormal behaviors, then detection coverage is improved, but system complexity increases

Engineering Contradiction:
Improvedetection coverageVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges multiple detection methods into a unified graph embedding framework. Different abnormal behaviors (phishing, Ponzi schemes, money laundering, terrorist financing) are detected through a single integrated system that processes transaction data uniformly, eliminating the need for separate detection modules for each behavior type.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The graph embedding model serves as a universal detection mechanism that can identify multiple types of abnormal behaviors simultaneously. The same core algorithm processes diverse transaction patterns, making the system multi-functional without requiring separate specialized detectors for each crime type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Adaptability or versatility

If multiple detection methods are integrated to detect all possible abnormal behaviors, then detection coverage is improved, but detection stability decreases

Engineering Contradiction:
Improvedetection coverageVSAvoiddetection stability
Core Design Contradiction:
Adaptability or versatilityVSReliability

Solution Approach 1:

By combining multiple detection approaches into a single graph embedding system, the patent eliminates the instability caused by integrating disparate methods. The unified architecture ensures consistent performance across different abnormal behavior types, improving overall detection stability while maintaining comprehensive coverage.

Inventive Principle:
Principle #5Merging (Combining)

3Adaptability or versatility

If multiple detection methods are integrated to detect all possible abnormal behaviors, then detection coverage is improved, but processing time increases

Engineering Contradiction:
Improvedetection coverageVSAvoidprocessing time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The graph embedding approach allows the system to skip the time-consuming process of sequentially applying multiple detection methods. By processing all abnormal behavior detection through a single unified model, the system rushes through the detection process efficiently, reducing processing time while maintaining comprehensive coverage.

Inventive Principle:
Principle #21Skipping (Rushing through)

4Adaptability or versatility

If multiple detection methods are integrated to detect all possible abnormal behaviors, then detection coverage is improved, but server burden increases

Engineering Contradiction:
Improvedetection coverageVSAvoidserver burden
Core Design Contradiction:
Adaptability or versatilityVSUse of energy by moving object

Solution Approach 1:

The patent merges multiple detection functions into a single server process running the graph embedding model. This consolidation reduces the server burden compared to running multiple separate detection systems, as the unified model processes all abnormal behavior types in one pass rather than requiring multiple independent processing streams.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11803855B2Method for detecting block chain abnormal behavior based on graph embedding
Publication Date: 2023.10.31 YANTAI UNIV
  • US11803855B2 patent drawing

AI summary

Disclosed is a method for detecting a block chain abnormal behavior based on graph embedding. The method comprises S100: data collection: acquiring public block chain abnormal behavior node data on the Internet, and acquiring normal nodes in a number equal to that of abnormal behavior nodes at the same time; S200: establishment of abnormal behavior recognition model: extracting features of all nodes, constructing the nodes subjected to feature extraction into a transaction graph, and forming the abnormal behavior recognition model based on a graph embedding technology according to the constructed transaction graph; and S300: transaction detection: determining a transaction risk according to the obtained abnormal behavior recognition model when a transaction occurs, and prompting a user of a risk level. According to the method for detecting the block chain abnormal behavior based on graph embedding, abnormal behaviors possibly existing in block chain transactions can be effectively detected and early warned.