Implicit Concept Drift Detection for Multi-label Data Streams

TL;DR

This paper introduces LD3, an unsupervised algorithm that detects concept drift in multi-label data streams by analyzing label dependencies, significantly improving predictive performance over existing methods.

Contribution

The paper presents LD3, the first unsupervised concept drift detector tailored for multi-label data streams, leveraging label dependencies for improved detection accuracy.

Findings

LD3 outperforms 14 existing detectors by up to 68.6% in predictive performance.

LD3 is effective on both real-world and synthetic data streams.

The approach is unsupervised, reducing reliance on labeled data for drift detection.

Abstract

Many real-world applications adopt multi-label data streams as the need for algorithms to deal with rapidly changing data increases. Changes in data distribution, also known as concept drift, cause the existing classification models to rapidly lose their effectiveness. To assist the classifiers, we propose a novel algorithm called Label Dependency Drift Detector (LD3), an implicit (unsupervised) concept drift detector using label dependencies within the data for multi-label data streams. Our study exploits the dynamic temporal dependencies between labels using a label influence ranking method, which leverages a data fusion algorithm and uses the produced ranking to detect concept drift. LD3 is the first unsupervised concept drift detection algorithm in the multi-label classification problem area. In this study, we perform an extensive evaluation of LD3 by comparing it with 14 prevalent supervised concept drift detection algorithms that we adapt to the problem area using 12 datasets and a baseline classifier. The results show that LD3 provides between 19.8\% and 68.6\% better predictive performance than comparable detectors on both real-world and synthetic data streams.