Neural Network Distributed Representations for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting fraudulent electronic transactions are imperfect, often missing actual fraud while flagging innocuous transactions, leading to insecure data management and increased computer system loads.
Innovation Solution
A neural network system generates distributed representations of electronic transaction data, using unlabeled data to authorize transactions, reducing the cost of fraudulent transactions by improving performance over conventional models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If simple models and rules are used for fraud detection, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent transforms the fraud detection approach by changing the parameter representation from simple binary flags to distributed vector representations. Each entity (user, device, location) is represented as a high-dimensional vector that captures nuanced patterns from historical data, enabling more precise fraud detection without proportionally increasing system complexity.
Solution Approach 2:
The patent introduces an intermediary embedding layer that maps discrete entity identifiers into continuous vector spaces. This intermediary representation allows the system to capture complex relationships and patterns indirectly, improving detection precision while maintaining relatively simple model architecture.
2Productivity
If conventional fraud detection methods are used, then processing speed is maintained, but reliability deteriorates
Solution Approach 1:
The patent pre-computes and stores distributed representations of entities (users, devices, locations) during off-peak periods. When a transaction occurs, the system quickly retrieves and compares these pre-computed vectors rather than performing complex computations in real-time, thus maintaining processing speed while improving detection reliability through more sophisticated analysis.
3Measurement precision
If more data is processed to improve fraud detection, then measurement precision improves, but use of energy increases
Solution Approach 1:
The patent creates compressed vector representations (embeddings) that capture the essential characteristics of entities and their relationships. These compact vector copies allow the system to process and analyze data efficiently without requiring the full complexity of the original datasets, reducing computational energy consumption while maintaining detection accuracy.
Data Source
AI summary
Systems and methods are provided for authorizing an electronic transaction. In one implementation at least one processor is programmed to receive electronic transaction data and historical transaction data, the electronic transaction data including an entity identifier component and an amount component of an electronic transaction; determine, based on the entity identifier component and the amount component, a location of the electronic transaction in a space of a distributed representation space, the distributed representation space comprising a mapping of electronic transaction components in a high-order space; determine locations of the historical transaction data in the distributed representation space; determine a decision boundary in the distributed representation space based on the locations of the historical transaction data; and authorize the electronic transaction based on the location of the electronic transaction being within the decision boundary.