Multi-Class Classifier for Intercepting Erroneous Electronic Transactions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Electronic banking systems face challenges in automating and simplifying the review and correction of incorrect electronic transactions, such as bill payments, due to inconsistencies and user-input errors, leading to time-consuming and costly errors that can impact corporate creditors' ability to reconcile accounts and customer satisfaction.
Innovation Solution
A computer-implemented system that uses a multi-class classifier and machine learning to intercept potentially erroneous electronic transaction data processing tasks by analyzing parameters and user profile data, generating classification probability outputs, and preventing task execution if thresholds are not met, while also extracting features like time indicators and pattern indicators to determine incorrect characteristics of financial interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual review and correction of electronic transactions is performed, then error detection capability is improved, but processing time and operational costs increase
Solution Approach 1:
The patent replaces manual mechanical review processes with an automated machine learning system that uses classifiers and pattern recognition algorithms to detect errors in electronic transactions, thereby eliminating the need for human operators while maintaining high detection accuracy
Solution Approach 2:
The system enables self-service error detection by automatically analyzing transaction data, identifying anomalies, and flagging potential errors without requiring external human intervention, allowing the system to correct its own operational inefficiencies
2Productivity
If automated transaction processing is implemented, then processing speed is improved, but error rate increases due to user-input inconsistencies
Solution Approach 1:
The system performs preliminary error detection and validation before transactions are fully processed by analyzing data patterns, checking for inconsistencies, and flagging suspicious transactions in advance, preventing errors from propagating through the automated processing system
Solution Approach 2:
The patent implements feedback mechanisms where the machine learning system continuously learns from detected errors and user corrections, refining its classification models to improve accuracy over time and reduce the error rate in automated processing
3Measurement precision
If comprehensive data validation is performed, then transaction accuracy is improved, but system complexity increases
Solution Approach 1:
The patent segments the validation process into multiple specialized components including data preprocessing modules, feature extraction units, classification algorithms, and error reporting mechanisms, allowing each component to focus on specific validation tasks and reducing overall system complexity through modular design
Data Source
AI summary
Embodiments relate to systems and methods for intercepting potentially erroneous electronic transaction data processing tasks. The system includes a memory and processor configured for: receiving a processing task data set including at least one parameter for executing an electronic transaction data processing task; providing, to a multi-class classifier, an input data set with the at least one parameter for executing the electronic transaction data processing task and at least one data feature associated with a user profile to generate a classification probability output for each class in the multi-class classifier; and when none of the classification probability outputs meet a threshold condition, preventing execution of the electronic transaction data processing task.