Detection of Abnormal Input-Output Associations

TL;DR

This paper introduces a new outlier detection method that identifies abnormal input-output associations by analyzing data in the conditional relation space using a probabilistic model, effectively detecting multivariate outliers.

Contribution

The paper proposes a novel approach for detecting multivariate conditional outliers by modeling data in the input-output relation space with a decomposable probabilistic framework.

Findings

Successfully identifies multivariate conditional outliers

Demonstrates effectiveness on experimental datasets

Outperforms existing outlier detection methods

Abstract

We study a novel outlier detection problem that aims to identify abnormal input-output associations in data, whose instances consist of multi-dimensional input (context) and output (responses) pairs. We present our approach that works by analyzing data in the conditional (input--output) relation space, captured by a decomposable probabilistic model. Experimental results demonstrate the ability of our approach in identifying multivariate conditional outliers.