Machine Learning Algorithm for Detecting Fraudulent Recurring Transactions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current financial accounting systems lack the capability to detect fraudulent recurring charges, leading to customers unknowingly having such charges on their accounts for extended periods, which can significantly affect their financial stability.
Innovation Solution
A system and method utilizing a machine-learning algorithm that analyzes transaction data, customer feedback, and merchant information to identify suspicious recurring charges, including communication with external entities to verify merchant reputation and customer complaints, and notifies customers or terminates suspicious transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current financial accounting systems are used to track transactions, then basic transaction recording is maintained, but the system cannot detect fraudulent recurring charges
Solution Approach 1:
The patent introduces an intermediary machine learning algorithm that acts as a mediator between the existing financial accounting system and the detection of fraudulent recurring charges. This algorithm analyzes transaction data, customer feedback, and merchant information to identify suspicious patterns without requiring fundamental changes to the core accounting system architecture.
Solution Approach 2:
The patent replaces manual accounting mechanisms with an automated machine learning system. Instead of relying on traditional accounting methods that cannot detect recurring charges, the system uses algorithms that automatically analyze transaction patterns, customer communications, and merchant data to identify fraudulent activities.
2Ease of operation
If recurring charges are allowed to continue without detection, then customer authorization is simplified, but customers remain unaware of fraudulent charges for extended periods
Solution Approach 1:
The patent implements preliminary detection actions by continuously monitoring transactions against established patterns of legitimate recurring charges. The system proactively identifies suspicious charges before they cause significant financial harm to customers, analyzing each transaction in real-time against historical data and customer profiles.
Solution Approach 2:
The patent establishes a feedback mechanism where the system continuously learns from customer interactions, complaints, and resolved fraudulent charges. This feedback loop enables the machine learning algorithm to improve its detection accuracy over time, adjusting its patterns recognition based on actual customer behavior and fraud resolution data.
3Measurement precision
If the system monitors all transactions for recurring patterns, then detection accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent applies local quality by tailoring the detection analysis to specific transaction characteristics and customer profiles. Rather than applying uniform analysis to all transactions, the system adjusts its monitoring intensity and analysis depth based on the specific patterns, amounts, and customer histories, focusing computational resources where they are most needed.
Solution Approach 2:
The patent utilizes parameter changes by dynamically adjusting detection thresholds and analysis parameters based on learned patterns and fraud statistics. The machine learning algorithm continuously refines its detection parameters based on emerging fraud patterns and resolution data, optimizing the balance between detection accuracy and processing efficiency.
Data Source
AI summary
A system and method for detecting a recurring charge that is unknown to the customer and/or the result of fraud or deceit is disclosed. A customer complaint initiates an analysis of a particular transaction. In response to the complaint, information relating to the transaction is extracted and used to identify additional information from a variety of different sources. Such information may include information relating to the customer, information relating to the merchant, as well as information relating to past transactions or related transactions. This information is provided to a machine learning algorithm, both for training purposes and for analysis purposes. The machine learning algorithm analyzes a transaction in order to determine the likelihood that the transaction is a recurring transaction, as well as the likelihood that it is a result of a bad actor, fraud or deceit.