Capital one services, llc (20240346505). SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON BALANCED DATA simplified abstract
SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON BALANCED DATA
Organization Name
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 BALANCED DATA - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240346505 titled 'SYSTEMS AND METHODS FOR OUTLIER DETECTION USING UNSUPERVISED MACHINE LEARNING MODELS TRAINED ON BALANCED DATA
The abstract describes methods and systems for outlier detection, including the use of a single-tier or two-tier machine learning model.
- The system may generate a minority expert model by training an unsupervised machine learning model on oversampled training data.
- The minority expert model may provide a binary result indicating if an event is an outlier or not, or a multi-class result indicating sub-categories of outliers.
- The system can perform outlier detection on 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 in anomaly detection - Enhancing decision-making processes
Benefits: - Increased efficiency in outlier detection - Reduction in false positives - Enhanced data security and integrity
Commercial Applications: Title: Advanced Outlier Detection System for Enhanced Security Measures This technology can be used in various industries such as finance, healthcare, and e-commerce to improve fraud detection and anomaly identification processes, leading to better overall security measures and risk management strategies.
Prior Art: Researchers in the field of machine learning and data analytics have explored various techniques for outlier detection, including ensemble methods and deep learning algorithms. It is essential to review existing literature and patents related to outlier detection to understand the current state of the art in this field.
Frequently Updated Research: Stay informed about the latest advancements in machine learning algorithms and outlier detection techniques to ensure the continued relevance and effectiveness of this technology in various applications.
Questions about Outlier Detection: 1. How does the system differentiate between normal events and outliers in the dataset? The system uses a minority expert model trained on oversampled data to identify outliers based on predefined criteria. 2. Can the system adapt to new types of outliers that were not present in the training data? The system's machine learning models can be retrained periodically to incorporate new outlier patterns and improve detection accuracy over time.
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.
