Detecting Unusual Input-Output Associations in Multivariate Conditional Data

TL;DR

This paper introduces a novel framework for detecting multivariate conditional outliers by modeling input-output associations with a probabilistic approach, effectively identifying unusual responses in context-dependent data.

Contribution

It proposes a decomposable probabilistic model with weighted components to detect conditional outliers, addressing limitations of existing unconditional outlier detection methods.

Findings

Successfully identifies multivariate conditional outliers across various domains.

Demonstrates the effectiveness of global and local component weighting schemes.

Outperforms traditional methods in detecting context-dependent anomalies.

Abstract

Despite tremendous progress in outlier detection research in recent years, the majority of existing methods are designed only to detect unconditional outliers that correspond to unusual data patterns expressed in the joint space of all data attributes. Such methods are not applicable when we seek to detect conditional outliers that reflect unusual responses associated with a given context or condition. This work focuses on multivariate conditional outlier detection, a special type of the conditional outlier detection problem, where data instances consist of multi-dimensional input (context) and output (responses) pairs. We present a novel outlier detection framework that identifies abnormal input-output associations in data with the help of a decomposable conditional probabilistic model that is learned from all data instances. Since components of this model can vary in their quality, we combine them with the help of weights reflecting their reliability in assessment of outliers. We study two ways of calculating the component weights: global that relies on all data, and local that relies only on instances similar to the target instance. Experimental results on data from various domains demonstrate the ability of our framework to successfully identify multivariate conditional outliers.