Mobile App Anomaly Resolution Using ML-Guided Transaction Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for resolving suspicious account activity, such as unauthorized transactions, often require significant user delay and resource-intensive customer service interactions, leading to potential further unauthorized transactions before issues are resolved.
Innovation Solution
A mobile application leveraging biometric login and a machine learning model to present targeted questions to users for immediate resolution of suspicious activity, reducing the need for customer service representative intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a customer service representative is contacted to resolve suspicious activity, then the issue can be investigated and resolved, but significant time delay occurs and additional unauthorized transactions may happen
Solution Approach 1:
The system enables users to resolve suspicious activity independently through automated interactions with an AI assistant. Users receive push notifications about suspicious transactions and can confirm or deny them directly through the mobile application, eliminating the need to contact customer service representatives and significantly reducing resolution time while maintaining reliability through automated verification processes
Solution Approach 2:
The system performs preliminary verification of suspicious transactions by automatically analyzing transaction data, user behavior patterns, and device information before user confirmation. This preliminary action filters out obvious legitimate transactions and prepares verification questions in advance, enabling faster user response and reducing overall resolution time
2Reliability
If traditional customer service systems are used, then comprehensive verification can be performed, but resource-intensive manual intervention is required
Solution Approach 1:
The system replaces manual customer service representative intervention with an automated AI assistant that uses machine learning models to analyze transaction patterns, user behavior, and device characteristics. This substitution maintains verification accuracy through sophisticated automated analysis while dramatically improving productivity by eliminating manual processing bottlenecks
Solution Approach 2:
The system implements automated feedback loops where user responses to verification questions immediately update the risk assessment and transaction status. The AI assistant continuously processes user inputs, compares them against learned patterns, and automatically adjusts verification requirements, enabling efficient automated decision-making without manual intervention
Data Source
AI summary
A method may include determining suspicious activity with respect to a user account, the suspicious activity including at least one transaction on the user account; transmitting a notification to a computing device associated with the user account, the notification identifying the suspicious activity and including a link into an application installed on the computing device to confirm the suspicious activity; receiving, from the application, an indication that the notification was activated on the computing device; and in response to the indication: selecting a subset of queries of a plurality of queries to present on a display device of the computing device via the application; receiving answers to the subset of queries via the application; processing the answers to generate a resolved status of the suspicious activity.