(1 + ε)-class Classification: an Anomaly Detection Method for Highly Imbalanced or Incomplete Data Sets

06/14/2019
by   Maxim Borisyak, et al.
33

Anomaly detection is not an easy problem since distribution of anomalous samples is unknown a priori. We explore a novel method that gives a trade-off possibility between one-class and two-class approaches, and leads to a better performance on anomaly detection problems with small or non-representative anomalous samples. The method is evaluated using several data sets and compared to a set of conventional one-class and two-class approaches.

READ FULL TEXT

page 4

page 8

page 9

research
09/06/2019

Image anomaly detection with capsule networks and imbalanced datasets

Image anomaly detection consists in finding images with anomalous, unusu...
research
05/12/2016

Detecting Relative Anomaly

System states that are anomalous from the perspective of a domain expert...
research
02/28/2015

Sequential Feature Explanations for Anomaly Detection

In many applications, an anomaly detection system presents the most anom...
research
04/19/2022

"Flux+Mutability": A Conditional Generative Approach to One-Class Classification and Anomaly Detection

Anomaly Detection is becoming increasingly popular within the experiment...
research
02/01/2019

Automatic Hyperparameter Tuning Method for Local Outlier Factor, with Applications to Anomaly Detection

In recent years, there have been many practical applications of anomaly ...
research
04/25/2018

Putin's peaks: Russian election data revisited

We study the anomalous prevalence of integer percentages in the last par...
research
05/17/2021

DOC3-Deep One Class Classification using Contradictions

This paper introduces the notion of learning from contradictions (a.k.a ...

Please sign up or login with your details

Forgot password? Click here to reset