An adaptive classification framework for unsupervised model updating in nonstationary environments

Piero Conca*, Jon Timmis, Rogério De Lemos, Simon Forrest, Heather McCracken

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference Proceeding (Non-Journal item)

3 Citations (Scopus)

Abstract

This paper introduces an adaptive framework that makes use of ensemble classification and self-training to maintain high classification performance in datasets affected by concept drift without the aid of external supervision to update the model of a classifier. The updating of the model of the framework is triggered by a mechanism that infers the presence of concept drift based on the analysis of the differences between the outputs of the different classifiers. In order to evaluate the performance of the proposed algorithm, comparisons were made with a set of unsupervised classification techniques and drift detection techniques. The results show that the framework is able to react more promptly to performance degradation than the existing methods and this leads to increased classification accuracy. In addition, the framework stores a smaller amount of instances with respect to a single-classifier approach.

Original languageEnglish
Title of host publicationMachine Learning, Optimization, and Big Data - 1st International Workshop, MOD 2015 Taormina, Revised Selected Papers
EditorsMario Pavone, Giovanni Maria Farinella, Vincenzo Cutello, Panos Pardalos
PublisherSpringer Nature
Pages171-184
Number of pages14
ISBN (Print)9783319279251
DOIs
Publication statusPublished - 2015
Event1st International Workshop on Machine Learning, Optimization, and Big Data, MOD 2015 - Taormina, Sicily, Italy
Duration: 21 Jul 201523 Jul 2015

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9432
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference1st International Workshop on Machine Learning, Optimization, and Big Data, MOD 2015
Country/TerritoryItaly
CityTaormina, Sicily
Period21 Jul 201523 Jul 2015

Fingerprint

Dive into the research topics of 'An adaptive classification framework for unsupervised model updating in nonstationary environments'. Together they form a unique fingerprint.

Cite this