Anomalous Payment Detection and Intervention System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face difficulties in detecting and resolving unintended or erroneous payments, leading to unnecessary overdraft fees and time-consuming processes, as they must manually correct errors that may not be their own, without immediate access to their funds.
Innovation Solution
A method implemented by a provider computing system that generates a set of payments to identify anomalies by comparing actual and predicted categorizations, flagging payments as anomalous and problematic based on user behavior and experiences, and transmitting alerts to initiate proactive interventions such as calls or notifications to resolve issues before they escalate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automated payments and mobile applications are used to make payments easier, then payment convenience is improved, but the complexity of detecting and resolving payment errors increases
Solution Approach 1:
The system automatically detects anomalous payments by comparing actual payments against predicted payment patterns, and autonomously initiates interventions without requiring user action. The payment system serves itself by identifying and resolving errors that would otherwise require manual user detection and correction.
Solution Approach 2:
The system implements feedback loops where payment anomalies are detected, alerts are transmitted to intermediary devices, and interventions are executed based on the feedback from the anomaly detection. This closed-loop feedback mechanism automatically resolves payment errors without increasing user burden.
2Measurement precision
If users manually monitor and correct payment errors, then payment accuracy is improved, but the time and effort required increases significantly
Solution Approach 1:
The system performs preliminary actions by detecting and flagging anomalous payments before they result in significant financial harm. By identifying payment anomalies early in the payment process, the system prevents errors from escalating and reduces the time users would otherwise spend correcting them.
Solution Approach 2:
The system replaces manual user monitoring and correction actions with automated computational analysis. Machine learning models predict expected payments and automatically identify anomalies, substituting the mechanical process of manual review with automated algorithmic detection that operates continuously without user time investment.
3Adaptability or versatility
If automated draft termination is implemented, then user control over payments is improved, but the reliability of payment processing decreases
Solution Approach 1:
The system introduces an intermediary anomaly detection layer between the automatic draft termination request and the payment processing system. This intermediary monitors the termination process, detects anomalies such as duplicate drafts or processing failures, and triggers corrective interventions, thereby maintaining reliability while preserving user control.
Solution Approach 2:
The system prepares cushioning measures by having pre-configured intervention protocols ready to activate when draft termination anomalies are detected. This beforehand preparation ensures that reliability issues are quickly addressed without compromising user control, as the corrective actions are already planned and ready to execute.
4Reliability
If insufficient funds checks are performed, then overdraft prevention is improved, but the productivity of account operations decreases
Solution Approach 1:
The system performs preliminary anomaly detection on payments before they are fully processed, identifying potential overdraft situations early. By detecting anomalous payments that could cause insufficient funds issues before execution, the system prevents overdrafts without requiring time-consuming manual checks that would reduce operational productivity.
Data Source
AI summary
Disclosed are methods, systems, and devices for identifying anomalous and/or problematic payments based on past payment patterns of a user, past experiences with the user, and the user's circumstances, and implementing intervention mechanisms for proactively resolving account issues before downstream impacts, such as collections calls, inbound calls, and complaints. Anomalous payments may be detected using clustering and prediction. Interventions may include phone calls, emails, notifications, etc. An alert or notification may be sent to the user's mobile application, and various selectable response options suited to the situation may be presented to the user.


