Diversity Measure for Drift Detection in Data Streams

EasyChair Preprint 15689

Abstract

Concept drift is a notable challenge in machine learning, data mining and applications involving big data and large-scale data processing. The employment of diversity measures has emerged as an effective strategy. We have examined and investigated the role of the diversity measure in detecting concept drift and provided comparative analysis of four different ways of using them: DMDDM for drift detection in a fully supervised binary classification context, DMDDM-S in a semi-supervised context, DMODD for online drift detection in a fully supervised multi-classification context, and HBBE, a hybrid block-based ensemble designed for addressing different types of concept drifts. Our comparative analysis evaluates the efficacy of these methods in detecting concept drift and enhancing model performance. The results confirm the effectiveness of all four approaches within their respective settings. Overall, this paper accentuates the significance of diversity measures in addressing concept drift and outlines a trajectory for their sustained evolution in machine /data stream learning contexts.

Keyphrases: Concept Drifts, DMDDM, DMODD, Data Mining, Drift detection methods, HBBE, concept drift, concept drift and enhancing model, concept drift detection, data streams, detecting concept drift, diversity measure, diversity measures for concept drift detection, drift detection method, online drift detection, online drift detector

@booklet{EasyChair:15689,
  author    = {Osama A. Mahdi and Savitri Bevinakoppa and Sarabjot Singh},
  title     = {Diversity Measure for Drift Detection in Data Streams},
  howpublished = {EasyChair Preprint 15689},
  year      = {EasyChair, 2025}}