Channel Change Entropy Metrics for Streaming Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fraud analytics in digital content streaming are not granular enough to accurately detect credential sharing and network errors, and entropy-based methods are computationally expensive for large-scale implementations.
Innovation Solution
An automated system uses channel change sequence entropy metrics, specifically the Gini index, to determine channel diversity and identify anomalies by analyzing DRM license requests and user behavior patterns, leveraging machine learning to detect fraudulent usage patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If entropy-based anomaly detection methods are used to detect credential sharing and network errors, then detection accuracy is improved, but computational cost increases significantly for large-scale implementations
Solution Approach 1:
The patent extracts only the essential features needed for anomaly detection from the full channel change sequence data. Instead of computing entropy on complete sequences, the system identifies and extracts key characteristic patterns that suffice for detection, thereby reducing computational overhead while preserving detection accuracy.
Solution Approach 2:
The patent segments the channel change sequence data into smaller, manageable units for analysis. By dividing the data stream into discrete segments and analyzing each segment independently, the system reduces the computational burden of entropy calculation while maintaining the ability to detect anomalies across the entire data set.
2Device complexity
If aggregated trend analysis is used for fraud detection, then computational complexity is reduced, but detection granularity and accuracy are insufficient
Solution Approach 1:
The patent introduces a new dimension of analysis by applying entropy metrics to channel change sequence data. This adds a temporal and behavioral dimension to the analysis, moving beyond simple aggregated trends to capture the dynamic patterns of user behavior, thereby improving detection granularity without excessive computational complexity.
Solution Approach 2:
The patent changes the analytical parameter from aggregate counts to entropy-based diversity metrics. By transforming the data representation and using entropy as the key parameter, the system achieves finer detection granularity while keeping computational requirements manageable through efficient entropy calculation methods.
3Measurement precision
If detailed channel change sequence analysis is performed on tens of millions of devices, then detection accuracy is improved, but computational expense becomes prohibitive
Solution Approach 1:
The patent applies partial action by performing detailed entropy analysis only on suspicious or anomalous channel change sequences rather than uniformly analyzing all data from tens of millions of devices. This selective approach maintains high detection accuracy for problematic cases while dramatically improving overall processing efficiency.
Solution Approach 2:
The patent performs preliminary filtering and preprocessing of channel change sequence data to identify potentially anomalous patterns before applying computationally intensive entropy analysis. This preliminary action reduces the data volume requiring detailed analysis, thereby improving processing efficiency while preserving detection accuracy for genuine anomalies.
Data Source
AI summary
Methods and systems for automated detection of credential theft and network errors using channel change sequence entropy metrics. A method includes determining, by an automated channel change sequence detection system using an entropy-based method, a channel diversity value for channel change sequence data collected for a unique entry-point for streaming content from a service provider system, where the channel diversity value is a measure of the diversity of the channel change sequence data, reviewing, by the automated channel change sequence detection system using machine learning, at least the channel diversity value to determine an issue, and outputting, by the automated channel change sequence detection system to a service provider system component associated with the determined issue, an issue message to act on the determined issue.