Cluster Metric Sensitivity to Irrelevant Features

TL;DR

This paper examines how irrelevant features affect clustering performance, revealing that some metrics are more sensitive than others and suggesting better metrics for feature selection in unsupervised learning.

Contribution

It provides an empirical analysis of the impact of irrelevant features on clustering metrics, highlighting the sensitivity differences among popular evaluation scores.

Findings

ARI and NMI are resilient to Gaussian irrelevant features.

Silhouette and Davies-Bouldin scores are highly sensitive to irrelevant features.

Irrelevant features can cause significant score fluctuations, especially in certain metrics.

Abstract

Clustering algorithms are used extensively in data analysis for data exploration and discovery. Technological advancements lead to continually growth of data in terms of volume, dimensionality and complexity. This provides great opportunities in data analytics as the data can be interrogated for many different purposes. This however leads challenges, such as identification of relevant features for a given task. In supervised tasks, one can utilise a number of methods to optimise the input features for the task objective (e.g. classification accuracy). In unsupervised problems, such tools are not readily available, in part due to an inability to quantify feature relevance in unlabeled tasks. In this paper, we investigate the sensitivity of clustering performance noisy uncorrelated variables iteratively added to baseline datasets with well defined clusters. We show how different types of irrelevant variables can impact the outcome of a clustering result from $k$-means in different ways. We observe a resilience to very high proportions of irrelevant features for adjusted rand index (ARI) and normalised mutual information (NMI) when the irrelevant features are Gaussian distributed. For Uniformly distributed irrelevant features, we notice the resilience of ARI and NMI is dependent on the dimensionality of the data and exhibits tipping points between high scores and near zero. Our results show that the Silhouette Coefficient and the Davies-Bouldin score are the most sensitive to irrelevant added features exhibiting large changes in score for comparably low proportions of irrelevant features regardless of underlying distribution or data scaling. As such the Silhouette Coefficient and the Davies-Bouldin score are good candidates for optimising feature selection in unsupervised clustering tasks.