Contextual One-Class Classification in Data Streams

TL;DR

This paper introduces a framework for using contextual information to enhance one-class classifiers in data streams, addressing challenges of dynamic environments and demonstrating improved performance through experiments.

Contribution

It proposes three frameworks for incorporating context into one-class classification in data streams, a novel approach for dynamic environments.

Findings

Context-aware classifiers outperform non-contextual ones.

Experimental results on synthetic and benchmark data support the effectiveness.

Using context reduces false positives in data stream classification.

Abstract

In machine learning, the one-class classification problem occurs when training instances are only available from one class. It has been observed that making use of this class's structure, or its different contexts, may improve one-class classifier performance. Although this observation has been demonstrated for static data, a rigorous application of the idea within the data stream environment is lacking. To address this gap, we propose the use of context to guide one-class classifier learning in data streams, paying particular attention to the challenges presented by the dynamic learning environment. We present three frameworks that learn contexts and conduct experiments with synthetic and benchmark data streams. We conclude that the paradigm of contexts in data streams can be used to improve the performance of streaming one-class classifiers.