Recurring Transaction Classification Using Merchant-Level Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for identifying recurring financial transactions are inaccurate, leading to misclassification of charges, which can result in customer dissatisfaction and inefficient user experiences due to the need for manual intervention to resolve errors.
Innovation Solution
A computing platform uses a trained machine-learning model to analyze transaction data, applying pre-processing and post-processing logic to accurately classify transactions as recurring or non-recurring based on feature data, including transaction-level, merchant-level, and account-merchant-level information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual intervention is used to resolve misclassifications, then accuracy can be improved, but time consumption and operational complexity increase
Solution Approach 1:
The system implements feedback mechanisms where classification results are reviewed and corrected, with corrections fed back into the system to improve future classifications. This allows the system to learn from manual interventions and reduce the need for ongoing manual resolution.
Solution Approach 2:
The system performs self-correction through automated processes that identify and resolve misclassifications without requiring continuous manual intervention. The machine learning models automatically update based on new data and patterns, enabling the system to service itself.
2Productivity
If automated classification is used, then efficiency improves, but classification accuracy deteriorates
Solution Approach 1:
The system replaces manual mechanical classification processes with automated machine learning models that use complex algorithms to analyze transaction patterns. This substitution maintains high accuracy while dramatically improving processing efficiency and productivity.
Solution Approach 2:
The system dynamically adjusts classification parameters and model thresholds based on changing transaction patterns and data characteristics. This allows the automated system to adapt to new scenarios and maintain high accuracy across evolving financial landscapes.
3Measurement precision
If complex analysis methods are applied, then classification accuracy improves, but system complexity increases
Solution Approach 1:
The system divides the complex classification task into separate modular components, including data collection modules, feature extraction modules, classification modules, and review modules. This segmentation allows each component to be optimized independently and simplifies the overall system architecture.
Solution Approach 2:
The system introduces intermediary layers such as feature extraction and data preparation components that bridge raw transaction data and final classification decisions. These intermediaries simplify the relationship between input data and output decisions, making the complex system more manageable and maintainable.
Data Source
AI summary
An example computing platform is configured to: (a) obtain data for a given transaction involving a given customer account and a given merchant, (b) apply pre-processing logic to the obtained data for the given transaction and thereby derive feature data for the given transaction, (c) input the feature data for the given transaction into a trained machine-learning model that functions to (i) evaluate the feature data for the given transaction and (ii) based on the evaluation, output a score for the given transaction that indicates a likelihood that the given transaction is a recurring charge, and (d) based on the score for the given transaction, determine whether to classify the given transaction as a recurring charge.