On Metafeatures' Ability of Implicit Concept Identification
Abstract
Concept drift in data stream processing remains an intriguing challenge and states a popular research topic. Methods that actively process data streams usually employ drift detectors, whose performance is often based on monitoring the variability of different stream properties. This publication provides an overview and analysis of metafeatures variability describing data streams with concept drifts. Five experiments conducted on synthetic, semi-synthetic, and real-world data streams examine the ability of over 160 metafeatures from 9 categories to recognize concepts in non-stationary data streams. The work reveals the distinctions in the considered sources of streams and specifies 17 metafeatures with a high ability of concept identification.
Cite
Text
Komorniczak and Ksieniewicz. "On Metafeatures' Ability of Implicit Concept Identification." Machine Learning, 2024. doi:10.1007/S10994-024-06612-0Markdown
[Komorniczak and Ksieniewicz. "On Metafeatures' Ability of Implicit Concept Identification." Machine Learning, 2024.](https://mlanthology.org/mlj/2024/komorniczak2024mlj-metafeatures/) doi:10.1007/S10994-024-06612-0BibTeX
@article{komorniczak2024mlj-metafeatures,
title = {{On Metafeatures' Ability of Implicit Concept Identification}},
author = {Komorniczak, Joanna and Ksieniewicz, Pawel},
journal = {Machine Learning},
year = {2024},
pages = {7931-7966},
doi = {10.1007/S10994-024-06612-0},
volume = {113},
url = {https://mlanthology.org/mlj/2024/komorniczak2024mlj-metafeatures/}
}