Deferred Transaction Fraud Detection via Segmented Modules
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing deferred transaction services face challenges in effectively reviewing requests for short-term financing services like Buy-Now-Pay-Later (BNPL) without prejudicing buyers or exposing sellers to unreasonable risk, while also needing to scale to meet demand and be portable for mobile transactions.
Innovation Solution
A system that includes a transaction strategy system with functional modules to review and selectively approve requests for deferred transactions, utilizing credential stuffing, synthetic identity theft, and triangulation modules to assess risk, and integrating with data sources for comprehensive analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If comprehensive fraud detection modules are implemented, then security against fraudulent activities is improved, but system complexity increases
Solution Approach 1:
The fraud detection system is divided into multiple specialized functional modules including credential stuffing module, synthetic identity theft module, account takeover module, trojan threat module, triangulation module, and chargeback fraud module. Each module independently handles a specific type of fraud detection, making the overall complex system manageable through functional segmentation while maintaining comprehensive security coverage
Solution Approach 2:
The patent introduces a transaction strategy system that acts as an intermediary layer between the user and the deferred transaction service provider. This intermediary coordinates the interactions between multiple fraud detection modules and manages the overall detection process, reducing the complexity burden on individual modules while maintaining comprehensive security
2Reliability
If multiple fraud detection modules are integrated, then fraud detection capability is improved, but processing time increases
Solution Approach 1:
The system performs preliminary fraud detection actions by implementing credential stuffing detection, synthetic identity theft detection, and account takeover detection before the actual transaction is processed. By detecting potential fraud indicators in advance through dedicated modules, the system prevents fraudulent transactions from proceeding to later stages, reducing overall processing time for legitimate transactions
Solution Approach 2:
The patent implements a multi-layered detection approach where not all fraud detection modules are activated for every transaction. Instead, the system applies detection methods proportionally based on transaction risk indicators, applying more comprehensive detection only when necessary. This partial action approach maintains fraud detection capability while optimizing processing time for low-risk transactions
Data Source
AI summary
A computer-implemented method for utilizing a machine learning model configured to determine synthetic identity theft may include processing a plurality of user datasets to generate a set of features for each user dataset, with each set of features being representative of a particular user. The method may further include generating a plurality of embeddings sets, with each embedding set being representative of a respective set of features, generating a plurality of synthetic user datasets, combining the plurality of embeddings sets and the plurality of synthetic user datasets to generate a training dataset, the training dataset comprising a plurality of user profiles, training the machine learning model based on the generated training dataset, and determining, via the machine learning model and in response to receiving a new user profile, a determination of whether the new user profile is real or synthetic.