October 2011 - January 2014
Also known as one-class classification, we set out to review methods and applications of novelty detection.
Project overview
Novelty detection is the task of classifying test data that differ in some respect from the data that are available during training. This may be seen as “one-class classification”, in which a model is constructed to describe “normal” training data. The novelty detection approach is typically used when the quantity of available “abnormal” data is insufficient to construct explicit models for non-normal classes. Applications include inference in datasets from critical systems, where the quantity of available normal data is very large, such that “normality” may be accurately modelled.
Download
Find the details for downloading the (Matlab) source code here.
Using the code should be straightforward: the download comes with some toy datasets on which the demo function may be used.
Relevant publications
[PDF]
A review of novelty detection
MAF Pimentel, DA Clifton, L Clifton, L Tarassenko
Signal Processing, 2014, 99, 215-49
Acknowledgements
A few people were involved on this work: Lei Clifton, David Clifton and Lionel Tarassenko. The project is also described here.