High-dimensional and large-scale anomaly detection using a linear one-class SVM with deep learning

作者：

Highlights：

• We use a combination of a one-class SVM and deep learning.

• In our model linear kernels can be used rather than nonlinear ones.

• Our model delivers a comparable accuracy with a deep autoencoder.

• Our model executes 3 times faster in training and 1000 faster than a deep autoencoder.

摘要

Highlights•We use a combination of a one-class SVM and deep learning.•In our model linear kernels can be used rather than nonlinear ones.•Our model delivers a comparable accuracy with a deep autoencoder.•Our model executes 3 times faster in training and 1000 faster than a deep autoencoder.

论文关键词：Anomaly detection,Outlier detection,High-dimensional data,Deep belief net,Deep learning,One-class SVM,Feature extraction

论文评审过程：Received 8 April 2015, Revised 29 February 2016, Accepted 28 March 2016, Available online 14 April 2016, Version of Record 26 May 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.03.028