Layered Data Overlay for Real-Time Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for real-time transaction processing face challenges in handling high volumes of data with millisecond latencies, managing siloed data, and providing accurate fraud detection amidst growing digital transaction volumes, with existing machine learning models struggling to adapt to real-world server constraints and security requirements.
Innovation Solution
The implementation of a machine learning system that dynamically updates models to detect anomalous transactions by storing and processing data in a layered state configuration, allowing for atomic data changes and overlay relationships between data sets, enabling fast inference and reducing processing latency while maintaining data integrity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional fraud detection rules are manually implemented in coded logic statements, then security control is established, but processing speed decreases and false positives increase
Solution Approach 1:
The patent replaces traditional mechanical rule-based fraud detection systems with machine learning models that can process transactions in real-time. The ML models analyze transaction patterns and detect fraud without requiring manual rule configuration, thereby maintaining security while dramatically improving processing speed and reducing false positives.
Solution Approach 2:
The patent changes the fundamental parameters of fraud detection by transitioning from static coded logic statements to dynamic machine learning models that continuously learn from transaction data. This allows the system to adapt to new fraud patterns while maintaining high processing speeds and accuracy.
2Measurement precision
If machine learning models are implemented for real-time transaction processing, then fraud detection accuracy improves, but system complexity and infrastructure requirements increase
Solution Approach 1:
The patent segments the machine learning system into distinct components: model training infrastructure, model deployment platform, and real-time inference engine. This segmentation allows each component to be optimized independently and simplifies the overall system architecture, making it more manageable despite the increased complexity of using ML models.
Solution Approach 2:
The patent introduces an intermediary layer between the machine learning models and the transaction processing system. This intermediary handles model deployment, monitoring, and updates, thereby simplifying the integration of complex ML models into the existing transaction infrastructure and reducing overall system complexity.
3Reliability
If data is stored in siloed or partitioned structures for security reasons, then data security is maintained, but data access and analysis efficiency decrease
Solution Approach 1:
The patent implements a universal data access layer that can efficiently query across multiple partitioned data stores while maintaining security protocols. This layer provides multi-functional access capabilities, allowing the system to maintain data security through partitioning while enabling fast data retrieval and analysis by abstracting away the complexity of data location and access paths.
Data Source
AI summary
A data item is searched for in first and/or second data sets of a data store. If it is found in the first data set and if it was updated in the first data set after the second data set became an overlay of the first data set, first data stored in association with the data item in the first data set is returned. If it is found in the first and second data sets and if the second data set became an overlay of the first data set after the data item was updated in the first data set, second data stored in association with the data item in the second data set is returned. The second data set is identified based on overlay metadata, indicative of the second data set being an overlay of the first data set.