User Flow Model for Fraudulent Access Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional online web services lack robust protection against unauthorized access, as they primarily rely on username and password verification, which is vulnerable to hacking and fails to monitor user activity post-authentication, allowing unauthorized users to access secure content without detection.
Innovation Solution
Implementing a system that monitors user activity and compares it to a model of normal behavior, generated based on prior interactions, to detect deviations and flag unauthorized access, using a remote computing device to track and analyze user interactions, including mouse movements, link selections, and data manipulation, and revoke access if suspicious activity is detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If username and password verification is used for security, then authentication is simple and quick, but security is vulnerable to hacking and unauthorized access
Solution Approach 1:
The system performs preliminary actions by continuously monitoring and recording user interaction data (mouse movements, clicks, scrolling) during normal authenticated sessions. This baseline behavior data is collected and stored before any unauthorized access attempt occurs, enabling the system to have security verification ready in advance rather than reacting after compromise.
Solution Approach 2:
The system implements feedback by continuously comparing real-time user interaction patterns against the established baseline model. When deviations are detected, the system provides feedback through security alerts and can trigger re-authentication requirements. This closed-loop feedback mechanism allows the system to dynamically respond to potential unauthorized access while maintaining normal operation during legitimate use.
2Reliability
If additional security mechanisms are implemented to detect fraudulent activity, then security is improved, but system complexity increases
Solution Approach 1:
The system applies self-service by automatically monitoring user behavior, comparing interactions against baseline models, and detecting anomalies without requiring manual security configuration or intervention. The system self-adjusts by continuously updating behavior models based on observed patterns, reducing the need for complex manual security rule maintenance while improving detection capabilities.
3Measurement precision
If user activity monitoring is implemented to detect unauthorized access, then detection capability is improved, but processing overhead and computational resources increase
Solution Approach 1:
The system applies partial action by selectively monitoring specific interaction elements (links, buttons, forms) rather than all possible user actions. The monitoring intensity is adjusted based on the sensitivity level assigned to different webpage elements, focusing computational resources on high-risk areas while reducing overhead for low-risk interactions. This selective approach maintains detection accuracy for critical security events while conserving computational resources.
Data Source
AI summary
Techniques to detect unauthorized access and/or use of a user's account provided by an online web service are provided. A user's activity on the online web service can be monitored to determine a model of the user flow through the online web service. The model of user flow can be unique to the user or can be a model assigned to a cluster of users having similar behaviors or mannerisms when navigating the online web service. The model can vary based on the type of device used by the user to access the online web service. Once a model is determined, subsequent activity on the online web service by the user can be monitored and compared to the model of user flow. Unauthorized access and/or use of the user's online account can be detected if the subsequent activity fails to fit the model of the user flow.