User Interaction Profile Correlation for Real-Time Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computer systems lack the ability to effectively correlate web-based activities with external activities, leading to incomplete understanding of user behaviors beyond graphical user interfaces, limiting real-time updates and authentication processes.
Innovation Solution
A method that involves generating data links between user interaction profiles and activity records, using tracking data from online interactions to predict correlation parameters, and updating profiles to enhance authentication and fraud detection by correlating online interactions with entity-related activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If systems track only GUI-based interactions, then tracking implementation is simple, but understanding of user behavior is incomplete
Solution Approach 1:
The patent introduces an intermediary data object structure that acts as a mediator between GUI tracking systems and external activity monitoring. This intermediary layer enables the system to capture and correlate both GUI-based interactions and external activities without fundamentally redesigning the entire tracking architecture, thus reducing information loss while managing complexity.
Solution Approach 2:
The tracking system is segmented into separate modules: one for capturing GUI interactions and another for monitoring external activities. These segmented components can independently collect data and then integrate through correlation mechanisms, allowing comprehensive behavior tracking while maintaining implementation simplicity through modular architecture.
2Reliability
If real-time updates are implemented, then authentication accuracy improves, but processing time increases
Solution Approach 1:
The system performs preliminary actions by pre-establishing correlation parameters and activity models before authentication is needed. Historical data is analyzed in advance to build predictive models that can quickly evaluate new interactions, enabling real-time authentication decisions without requiring complex real-time computations during the actual authentication process.
Solution Approach 2:
Real-time updates incorporate feedback mechanisms where system responses to user actions are immediately fed back into the correlation analysis. This feedback loop allows the system to adjust authentication decisions based on current behavior patterns, improving accuracy while the automated feedback processing occurs efficiently in the background without significant user-perceived delay.
3Reliability
If comprehensive behavior tracking is implemented, then fraud detection improves, but data processing complexity increases
Solution Approach 1:
The patent extracts and isolates specific correlation parameters from the vast amount of collected data. By identifying and separating the most relevant parameters that indicate fraudulent behavior, the system can process only the essential information needed for fraud detection rather than analyzing all available data, thus improving detection capability while reducing processing complexity.
Solution Approach 2:
The system transforms raw behavioral data into transformed correlation parameters through predefined models. By changing the representation of data from raw interaction logs to processed correlation parameters, the complexity of analyzing comprehensive behavior data is reduced while maintaining fraud detection accuracy, as the transformed parameters capture essential patterns more efficiently.
Data Source
AI summary
In order to provide improved matching of records between different sources, systems and methods include generating a data link between a stored interaction profile of the user and activity data records that identify activities performed by the user. Online interaction data associated with the user is received, including tracking data indicative of online interactions with content. The online interaction data is stored in the stored interaction profile associated with the user. An activity model is used to predict correlation parameters representing groupings of online interactions of the online interaction data with activities performed by the user, where the prediction is based on the tracking data and each activity in the interaction profile. The interaction profile is updated with the groupings and user activities are authenticated based on the interaction profile.