Synthetic Non-stationary Data Streams for Recognition of the Unknown

TL;DR

This paper introduces a synthetic data stream generation method that simulates both concept drifts and new class emergence, aiding in the development of open set recognition systems for non-stationary environments.

Contribution

It proposes a novel strategy for creating synthetic data streams that incorporate both concept drifts and unknown class emergence, facilitating research in open set recognition.

Findings

Unsupervised drift detectors effectively identify concept drifts and novelty.

Generated data streams improve open set recognition performance.

The approach enables testing of models in dynamic, non-stationary environments.

Abstract

The problem of data non-stationarity is commonly addressed in data stream processing. In a dynamic environment, methods should continuously be ready to analyze time-varying data -- hence, they should enable incremental training and respond to concept drifts. An equally important variability typical for non-stationary data stream environments is the emergence of new, previously unknown classes. Often, methods focus on one of these two phenomena -- detection of concept drifts or detection of novel classes -- while both difficulties can be observed in data streams. Additionally, concerning previously unknown observations, the topic of open set of classes has become particularly important in recent years, where the goal of methods is to efficiently classify within known classes and recognize objects outside the model competence. This article presents a strategy for synthetic data stream generation in which both concept drifts and the emergence of new classes representing unknown objects occur. The presented research shows how unsupervised drift detectors address the task of detecting novelty and concept drifts and demonstrates how the generated data streams can be utilized in the open set recognition task.