Machine Learning for Transaction Cleansing Override Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional systems for creating transaction cleansing overrides rely on manual human intervention, which is inefficient, time-consuming, and costly, and result in inaccurate merchant identification leading to false fraud reporting and incorrect spending limits.
Innovation Solution
A system utilizing machine learning models to analyze communications from cardholders, detect mislabeled merchant names, and generate merchant keys for automated creation of transaction cleansing overrides, reducing the need for manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual creation of transaction cleansing overrides is used, then human intervention ensures accuracy in merchant identification, but the process is inefficient, time-consuming, and costly
Solution Approach 1:
The patent replaces the manual mechanical process of creating transaction cleansing overrides with an automated machine learning-based system. The machine learning model analyzes merchant code data, detects mislabeled names, and generates cleansing overrides automatically, eliminating the need for human intervention while maintaining accuracy through algorithmic processing of transaction data
Solution Approach 2:
The system performs self-service by automatically detecting and correcting merchant identification errors without requiring human operators. The machine learning model autonomously processes communications from cardholders, identifies mislabeled names, and generates the necessary overrides independently, making the system self-sufficient and efficient
2Reliability
If manual overrides are created, then human review ensures correctness, but the process is time-consuming and costly
Solution Approach 1:
The patent substitutes manual human review with automated machine learning-based verification. The system analyzes communications from cardholders, detects mislabeled merchant names, and validates overrides through algorithmic processing, eliminating time-consuming manual review while maintaining high reliability through systematic verification processes
3Productivity
If merchant code data is used for merchant identification, then the system can detect merchants on financial statements, but inconsistent data leads to false fraud reporting and incorrect spending limits
Solution Approach 1:
The patent implements feedback mechanisms where the system receives communications from cardholders about mislabeled merchants, analyzes this feedback using machine learning models, and uses the results to correct merchant identification errors. This feedback loop continuously improves the accuracy of merchant name detection and ensures correct spending limit application
Solution Approach 2:
The system replaces simple mechanical merchant code data processing with advanced machine learning algorithms that analyze patterns in communications and transaction data. The machine learning model processes inconsistent merchant code data, detects mislabeled names, and generates accurate merchant identifiers, significantly improving measurement precision while maintaining detection capability
Data Source
AI summary
Disclosed embodiments may include a method for automated creation of transaction cleansing overrides. The system may include one or more processors, and memory in communication with the one or more processors and storing instructions that, when executed by the one or more processors, are configured to cause the system to receive a communication from a card holder via one of one or more communication channels and analyze the communication using a first machine learning model to determine an identity of the card holder, detect a mislabeled name, and detect a temporal indicator. In some embodiments, the mislabeled name and the temporal indicator are associated with a transaction. In some embodiments, the memory can be further configured to cause the system to detect card holder records and raw data associated with the transaction; generate a key based on the raw data; and alter an override list to add the key.