Modified Kernel Principal Component Analysis and Autoencoder Approaches to Unsupervised Anomaly Detection

Modified Kernel Principal Component Analysis and Autoencoder Approaches to Unsupervised Anomaly Detection

Unsupervised anomaly detection is the task of identifying examples that differ from the normal or expected pattern without the use of labeled training data. Our research addresses shortcomings in two existing anomaly detection algorithms, Kernel Principal Component Analysis (KPCA) and Autoencoders (...

Full description

Bibliographic Details
Main Author: Merrill, Nicholas Swede
Other Authors: Mechanical Engineering
Format: Others
Published: Virginia Tech 2020
Subjects:
Machine Learning
Deep Learning
Anomaly Detection
Autoencoder
Kernel Principal Component Analysis
Online Access:http://hdl.handle.net/10919/98659

Internet

http://hdl.handle.net/10919/98659

Similar Items