Fraud Detection System Using Multi-Phase Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional credit and debit card systems face significant challenges in detecting points of compromise, such as skimming and malware breaches, which require substantial computer resources due to the large number of transactions and data involved, leading to inefficient fraud detection and potential business losses.
Innovation Solution
The system employs tables or dataframes instead of matrices to process transaction data, using a multi-phase filtering process to identify potential points of compromise by calculating ratios and applying thresholds, thereby reducing computational resources and improving detection efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional matrix-based systems are used to detect points of compromise, then detection coverage can be maintained, but computational resources and processing time increase substantially
Solution Approach 1:
The patent segments the detection process into multiple phases: first identifying candidate locations using simplified criteria, then applying more rigorous analysis only to those candidates. This divides the large dataset into manageable portions for detailed analysis, reducing overall computational burden while maintaining detection reliability.
Solution Approach 2:
The patent extracts and focuses analysis only on the most promising candidates identified in earlier phases, rather than processing all transaction data with the same intensity. By taking out only the necessary subset for detailed investigation, the system reduces computational resource consumption while maintaining effective detection coverage.
2Measurement precision
If comprehensive transaction data is processed to detect points of compromise, then detection accuracy improves, but processing time increases
Solution Approach 1:
The patent performs preliminary filtering and candidate identification using lighter computational methods before applying rigorous analysis. This preliminary action narrows down the dataset to relevant candidates, ensuring that accurate detection is applied only where needed, thus maintaining detection accuracy while reducing overall processing time.
Solution Approach 2:
The patent employs dynamic analysis where the depth and complexity of data processing varies based on the characteristics of each candidate location. More resources are allocated to high-risk candidates that require detailed investigation, while low-risk areas receive minimal processing, optimizing the balance between accuracy and processing time.
3Reliability
If detailed analysis of all transactions is performed, then detection reliability improves, but system complexity increases
Solution Approach 1:
The patent segments the analysis into distinct phases with different complexity levels. Early phases use simple filtering rules that are easy to implement, while later phases apply more complex analysis only to candidates that passed preliminary filters. This segmentation maintains detection reliability through multi-stage verification while keeping overall system complexity manageable.
Solution Approach 2:
The patent applies partial analysis to the majority of transactions (using simplified methods) and reserves excessive/detailed analysis only for the small subset of candidates that warrant investigation. This partial application of complex analysis maintains reliability for critical cases while avoiding the complexity burden of analyzing all transactions with full rigor.
Data Source
AI summary
A system for maintaining data integrity includes a network interface and a processor coupled to memory. The processor can be configured to receive, via the network interface and from one or more computing devices, data regarding a plurality of transactions performed at a plurality of locations; generate, from the data, a data structure comprising a plurality of rows; for each of the plurality of locations, determine a ratio of a count of rows of a subset of rows for a location that each include an indication of a fraudulent transaction to a count of rows of the plurality of rows for the location; determine a location from the plurality of locations is a point of compromise based on the ratio for the location; and generate a record comprising a stored association between an identification of the location and an identification indicating a point of compromise responsive to the determination.