Media Fraud Detection Using Balanced ML Training Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Media services face challenges in detecting fraudulent user activity that manipulates media usage data, leading to inaccurate popularity charts, recommendations, and royalty payments, which erodes user trust and financial fairness.
Innovation Solution
A media fraud detection system using machine learning techniques, including a media fraud detection model trained on balanced training data, evaluates media usage data to generate fraud predictions, flagging abnormal behavior and adjusting metrics accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fraud detection methods are used, then the system is simple to implement, but the detection accuracy is insufficient to identify sophisticated fraudulent activity
Solution Approach 1:
The patent replaces traditional rule-based fraud detection mechanisms with machine learning models that automatically learn patterns from data. The system uses trained models to analyze media usage data and identify fraudulent behavior, substituting manual detection rules with adaptive algorithms that improve accuracy while managing complexity through automated training processes.
Solution Approach 2:
The system implements self-service through automated model training and evaluation processes. The fraud detection models are trained on historical data and automatically updated to adapt to new fraud patterns, reducing the need for manual intervention in system maintenance and improvement while enhancing detection capabilities.
2Measurement precision
If comprehensive media usage data is analyzed, then fraud detection accuracy improves, but processing time and computational resources increase
Solution Approach 1:
The patent applies preliminary action by pre-processing and preparing training data before model deployment. Historical media usage data is collected, labeled, and used to train fraud detection models in advance, so that when actual fraud detection is needed, the pre-trained models can quickly analyze new data without requiring extensive real-time processing.
Solution Approach 2:
The system segments the fraud detection process into distinct phases: data collection, model training, evaluation, and deployment. This segmentation allows computational-intensive tasks like model training to be performed separately from real-time detection, enabling comprehensive data analysis while maintaining fast response times during actual fraud identification.
3Reliability
If balanced training data is used, then model training effectiveness improves, but data preparation complexity increases
Solution Approach 1:
The patent extracts and separates the data balancing process as a distinct step in the workflow. Imbalanced training data is identified and corrected by extracting representative samples from underrepresented classes or synthesizing additional data, isolating the complexity of data preparation from the core model training process and enabling more effective and reliable model training.
Data Source
AI summary
Systems and techniques are disclosed for detecting fraudulent activity by users of a media service, such as a music or video streaming service. Fraudulent activity may artificially modify metrics tracked by the media service, thus impacting chart positions, media recommendations, royalty amounts, and other elements determined by the media service. Media usage data associated with user activity samples is labeled to identify fraud samples and non-fraud samples. Because the majority of user activity on the media service may be legitimate such a number of labeled non-fraud samples exceeds a number of labeled fraud samples, a balanced training data set is created via undersampling to reduce the number of non-fraud samples to be equal or closer to the number of fraud samples. The balanced training data set is used to train a machine learning model to generate fraud predictions associated with new user activity on the media service.