Double-Coupling Learning for Multi-Task Data Stream Classification

TL;DR

This paper introduces DC-SVM, a novel method for classifying multiple correlated data streams simultaneously by leveraging both inter-stream and intra-stream relationships, improving performance over traditional methods.

Contribution

The paper proposes a new double-coupling SVM approach that models correlations among multiple data streams for enhanced multi-task data stream classification.

Findings

DC-SVM outperforms traditional methods on artificial data.

The method shows improved accuracy on real-world multi-task data.

Experimental results validate the effectiveness of considering inter- and intra-stream correlations.

Abstract

Data stream classification methods demonstrate promising performance on a single data stream by exploring the cohesion in the data stream. However, multiple data streams that involve several correlated data streams are common in many practical scenarios, which can be viewed as multi-task data streams. Instead of handling them separately, it is beneficial to consider the correlations among the multi-task data streams for data stream modeling tasks. In this regard, a novel classification method called double-coupling support vector machines (DC-SVM), is proposed for classifying them simultaneously. DC-SVM considers the external correlations between multiple data streams, while handling the internal relationship within the individual data stream. Experimental results on artificial and real-world multi-task data streams demonstrate that the proposed method outperforms traditional data stream classification methods.