Describing Nonstationary Data Streams in Frequency Domain

TL;DR

This paper introduces the Frequency Filtering Metadescriptor, a novel tool for characterizing nonstationary data streams in the frequency domain to improve concept drift detection and understanding.

Contribution

It presents a new frequency-based approach for describing data streams, capable of identifying concepts and visualizing frequency components, outperforming existing strategies.

Findings

Effective in separating data chunks by concepts

Captures complex feature dependencies with fewer frequency components

Maintains semantic meaning in frequency domain

Abstract

Concept drift is among the primary challenges faced by the data stream processing methods. The drift detection strategies, designed to counteract the negative consequences of such changes, often rely on analyzing the problem metafeatures. This work presents the Frequency Filtering Metadescriptor -- a tool for characterizing the data stream that searches for the informative frequency components visible in the sample's feature vector. The frequencies are filtered according to their variance across all available data batches. The presented solution is capable of generating a metadescription of the data stream, separating chunks into groups describing specific concepts on its basis, and visualizing the frequencies in the original spatial domain. The experimental analysis compared the proposed solution with two state-of-the-art strategies and with the PCA baseline in the post-hoc concept identification task. The research is followed by the identification of concepts in the real-world data streams. The generalization in the frequency domain adapted in the proposed solution allows to capture the complex feature dependencies as a reduced number of frequency components, while maintaining the semantic meaning of data.