18300344. SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON OVERSAMPLED DATA simplified abstract (Capital One Services, LLC)

From WikiPatents
Revision as of 05:59, 18 October 2024 by Wikipatents (talk | contribs) (Creating a new page)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search

SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON OVERSAMPLED DATA

Organization Name

Capital One Services, LLC

Inventor(s)

Hassan Shallal of Plano TX (US)

Rajesh Kanna Durairaj of Plano TX (US)

SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON OVERSAMPLED DATA - A simplified explanation of the abstract

This abstract first appeared for US patent application 18300344 titled 'SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON OVERSAMPLED DATA

The abstract describes methods and systems for outlier detection, including a single-tier model with a minority expert model or a two-tier model with a majority expert model and a minority expert model. The system generates the minority expert model by training an unsupervised machine learning model on oversampled training data, including synthetic sample outlier events. The minority expert model may provide a binary result or a multi-class result indicating whether an event belongs to an outlier category or a sub-category of the outlier category.

  • Outlier detection system with single-tier or two-tier machine learning models
  • Minority expert model generated from unsupervised machine learning on oversampled data
  • Binary or multi-class results indicating outlier category or sub-category membership
  • Applicable to sequence-based or non-sequence-based events

Potential Applications: - Fraud detection in financial transactions - Anomaly detection in network security - Quality control in manufacturing processes

Problems Solved: - Identifying outliers in large datasets - Improving accuracy of outlier detection - Handling complex data distributions

Benefits: - Enhanced accuracy in outlier detection - Reduction in false positives - Improved decision-making based on outlier analysis

Commercial Applications: Title: Advanced Outlier Detection System for Enhanced Data Security This technology can be used in various industries such as finance, cybersecurity, and manufacturing to improve data security and quality control processes. The market implications include increased efficiency, reduced risks, and improved overall performance.

Questions about Outlier Detection: 1. How does the system handle outliers in real-time data streams? The system can be adapted to process real-time data streams by continuously updating the models and adjusting the outlier detection criteria based on the incoming data.

2. Can the system be integrated with existing data analytics platforms? Yes, the system is designed to be compatible with various data analytics platforms and can be easily integrated into existing systems for enhanced outlier detection capabilities.


Original Abstract Submitted

Methods and systems are described herein for outlier detection. The system may apply a single-tier including a minority expert model only or a two-tier machine learning model, including a majority expert model and a minority expert model. The system may generate the minority expert model by training an unsupervised machine learning model on oversampled training data, including synthetic sample outlier events. In some embodiments, the minority expert model may provide a binary result indicating an event belongs to an outlier category or not. In some embodiments, the minority expert model may include multiple component models providing a multi-class result indicating whether an event belongs to a sub-category of the outlier category. In application, the system may perform the outlier detection on events that are sequence-based or not.