Satyanarayanamurthy Mangipudi

You can explore some recent and related works here:

• https://www.sciencedirect.com/science/article/abs/pii/S1574013722000107
• https://www.sciencedirect.com/science/article/abs/pii/B9780323919197000202

• Article A hybrid dimensionality reduction method for outlier detecti...

• Article A Comparison of Outlier Detection Techniques for High-Dimensional Data

• Subspace Methods: Instead of examining the entire feature space, subspace methods focus on detecting outliers in lower-dimensional subspaces where anomalies are more apparent.
• Clustering-Based Approaches: Clustering methods can be adapted to identify clusters of normal data and then classify data points outside these clusters as outliers.
• Statistical Approaches: Techniques such as robust statistical measures (e.g., robust covariance estimation, robust Mahalanobis distance) are being developed to better handle high-dimensional data.
• Density-Based Methods: Density estimation techniques, such as local outlier factor (LOF) and density-based spatial clustering of applications with noise (DBSCAN), are extended to high-dimensional spaces.
• Ensemble Methods: Combining multiple outlier detection algorithms or using ensemble techniques can improve detection accuracy in high-dimensional scenarios.
• Deep Learning Approaches: Deep learning models, such as autoencoders, are being explored for anomaly detection in high-dimensional data, leveraging their ability to learn complex representations.
• Streaming and Online Methods: Algorithms that can adapt to streaming data or perform online outlier detection are becoming increasingly important with the rise of real-time data processing.