Identity Map for Abnormal Transaction Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing number of abnormal transactions in online markets, such as price manipulation and cross trading, poses challenges for detection due to users creating multiple accounts and conducting transactions within short time intervals to avoid detection, damaging brand reputations and causing disadvantages to consumers.
Innovation Solution
A method and apparatus that automatically detect abnormal transactions by analyzing transaction patterns, including unique and arbitrary information, access IP addresses, and generating an identity map to classify clusters and identify suspicious activity, utilizing a processor to extract identifiers and connect nodes for detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If users create multiple accounts to avoid detection, then the difficulty of detecting abnormal transactions increases, but the ability to conduct fraudulent transactions remains high
Solution Approach 1:
The patent combines multiple identifiers (user ID, IP address, device ID, transaction pattern) into a unified identity map that links multiple accounts to a single user. This merging approach allows the system to detect abnormal transactions even when users create multiple accounts, as all accounts are connected through the identity map based on shared identifying characteristics.
Solution Approach 2:
The patent introduces a new dimension of analysis by creating an identity map that operates alongside traditional account-based detection. This identity map adds a layer of indirect identification through IP addresses, device identifiers, and behavioral patterns, allowing detection without directly tracking account relationships.
2Reliability
If users conduct transactions within short time intervals, then the risk of price fluctuations is reduced, but the detectability of abnormal transactions decreases
Solution Approach 1:
The patent performs preliminary analysis by building the identity map in advance by collecting and correlating various identifying information (IP addresses, device IDs, user behaviors) before abnormal transactions occur. This pre-established identity map enables rapid detection of short-interval transactions, as the system already has the framework to link accounts and identify patterns without requiring time-consuming analysis during the transaction itself.
3Ease of manufacture
If traditional detection methods are used, then the implementation is simple, but the detection capability against sophisticated fraud is insufficient
Solution Approach 1:
The patent introduces an identity map as an intermediary structure that connects traditional detection methods with advanced fraud detection capabilities. The identity map serves as a mediator layer that aggregates identifying information and presents it in a format that enhances detection capability while building upon existing system infrastructure, thus maintaining relative implementation simplicity.
Data Source
AI summary
The present disclosure relates to a method of automatically detecting, by an electronic device, an abnormal transaction. The method may include: acquiring transaction information from an e-commerce server, the transaction information including unique information that is difficult for a user to arbitrarily change, arbitrary information that a user changes arbitrarily, and an access Internet protocol (IP) address; extracting a first identifier based on the unique information; extracting a plurality of second identifiers based on the arbitrary information; generating a third identifier based on the plurality of second identifiers; generating a first node based on the first identifier, and generating a second node based on the third identifier; generating a third node based on the access IP address; and connecting the first node, the second node, and the third node to generate an identity map for automatically detecting the abnormal transaction.