User Activity Similarity Metrics for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for tracking user activities in online digital platforms, particularly in educational settings, face challenges in analyzing and comparing user behavior across different user groups and integrating information from multiple web services with proprietary permissions, leading to limitations in identifying fraudulent activities and providing comprehensive user performance metrics.
Innovation Solution
A system and method for evaluating user activities in online digital environments that uses a content management system, a user evaluation web platform, and integrated data sources to calculate and display similarity metrics between users, enabling the identification of correct and fraudulent behaviors, and providing detailed performance insights without affecting proprietary permissions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If user activities are tracked across multiple web services with proprietary permissions, then comprehensive user behavior data is obtained, but data integration and analysis become more complex and difficult
Solution Approach 1:
The patent introduces an intermediary component that acts as a mediator between multiple web services with proprietary permissions and the analysis system. This intermediary collects activity records from various sources without requiring direct integration between them, thereby obtaining comprehensive user behavior data while avoiding the complexity of direct multi-service integration. The intermediary processes and standardizes data from different sources before passing it to the analysis module.
2Measurement precision
If detailed user activity information is collected and analyzed, then accurate performance metrics and fraud detection are achieved, but processing costs and computational resources increase
Solution Approach 1:
The patent extracts and analyzes only the most relevant features and patterns from the collected user activity data rather than processing all raw data. The system identifies key behavioral indicators and focuses computational resources on analyzing these extracted features, thereby achieving accurate performance metrics and fraud detection while significantly reducing processing costs and computational resource consumption.
3Reliability
If user activities are monitored across different user groups, then fraudulent activities can be identified, but privacy concerns and security risks increase
Solution Approach 1:
The patent segments user data into different groups and performs comparative analysis between segments rather than analyzing all user data in a single centralized system. This segmentation approach allows for effective fraud detection by comparing behaviors across user groups while maintaining better privacy protection, as individual user data remains within its segment and is not directly exposed to all other users' information.
Data Source
Figure 1~2
Figure 3
AI summary
The invention relates to a method and system for monitoring activities performed by user groups in online digital environments based on one or more web services comprising at least: one CMS (1) configured with different activities that can be performed by the users, and with one or more recording means for recording activity records (2) and/or results (3) associated with the performance of activities performed by said users, and a user evaluation web platform (7). Said platform (7) advantageously comprises a calculation module (9) configured for calculating the similarity of the activity records (2, 6, 6') of the CMS (1) and of the secondary data sources (5), evaluated between two or more users of the system, and associating said similarity with the result (3) obtained in the performance of said activities, and a display module (10) configured for reading and representing the calculated similarity information.