Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification
Most of the data stream classification methods need plenty of labeled samples to achieve a reasonable result. However, in a real data stream environment, it is crucial and expensive to obtain labeled samples, unlike the unlabeled ones. Although Active learning is one way to tackle this challenge, it...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Iran Telecom Research Center
2017-12-01
|
Series: | International Journal of Information and Communication Technology Research |
Subjects: | |
Online Access: | http://ijict.itrc.ac.ir/article-1-26-en.html |
_version_ | 1811169229351682048 |
---|---|
author | MohammadReza Keyvanpour Mahnoosh Kholghi Sogol Haghani |
author_facet | MohammadReza Keyvanpour Mahnoosh Kholghi Sogol Haghani |
author_sort | MohammadReza Keyvanpour |
collection | DOAJ |
description | Most of the data stream classification methods need plenty of labeled samples to achieve a reasonable result. However, in a real data stream environment, it is crucial and expensive to obtain labeled samples, unlike the unlabeled ones. Although Active learning is one way to tackle this challenge, it ignores the effect of unlabeled instances utilization that can help with strength supervised learning. This paper proposes a hybrid framework named “DSeSAL”, which combines active learning and dynamic self-training to achieve both strengths. Also, this framework introduces variance based self-training that uses minimal variance as a confidence measure. Since an early mistake by the base classifier in self-training can reinforce itself by generating incorrectly labeled data, especially in multi-class condition. A dynamic approach to avoid classifier accuracy deterioration, is considered. The other capability of the proposed framework is controlling the accuracy reduction by specifying a tolerance measure. To overcome data stream challenges, i.e., infinite length and evolving nature, we use the chunking method along with a classifier ensemble. A classifier is trained on each chunk and with previous classifiers form an ensemble of M such classifiers. Experimental results on synthetic and real-world data indicate the performance of the proposed framework in comparison with other approaches. |
first_indexed | 2024-04-10T16:40:03Z |
format | Article |
id | doaj.art-7443196057dd4e218380199bca1b4785 |
institution | Directory Open Access Journal |
issn | 2251-6107 2783-4425 |
language | English |
last_indexed | 2024-04-10T16:40:03Z |
publishDate | 2017-12-01 |
publisher | Iran Telecom Research Center |
record_format | Article |
series | International Journal of Information and Communication Technology Research |
spelling | doaj.art-7443196057dd4e218380199bca1b47852023-02-08T07:56:26ZengIran Telecom Research CenterInternational Journal of Information and Communication Technology Research2251-61072783-44252017-12-01943749Hybrid of Active Learning and Dynamic Self-Training for Data Stream ClassificationMohammadReza Keyvanpour0Mahnoosh Kholghi1Sogol Haghani2 Most of the data stream classification methods need plenty of labeled samples to achieve a reasonable result. However, in a real data stream environment, it is crucial and expensive to obtain labeled samples, unlike the unlabeled ones. Although Active learning is one way to tackle this challenge, it ignores the effect of unlabeled instances utilization that can help with strength supervised learning. This paper proposes a hybrid framework named “DSeSAL”, which combines active learning and dynamic self-training to achieve both strengths. Also, this framework introduces variance based self-training that uses minimal variance as a confidence measure. Since an early mistake by the base classifier in self-training can reinforce itself by generating incorrectly labeled data, especially in multi-class condition. A dynamic approach to avoid classifier accuracy deterioration, is considered. The other capability of the proposed framework is controlling the accuracy reduction by specifying a tolerance measure. To overcome data stream challenges, i.e., infinite length and evolving nature, we use the chunking method along with a classifier ensemble. A classifier is trained on each chunk and with previous classifiers form an ensemble of M such classifiers. Experimental results on synthetic and real-world data indicate the performance of the proposed framework in comparison with other approaches.http://ijict.itrc.ac.ir/article-1-26-en.htmlcomputer sciencedata miningsemi-supervised learningclassificationdata stream. |
spellingShingle | MohammadReza Keyvanpour Mahnoosh Kholghi Sogol Haghani Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification International Journal of Information and Communication Technology Research computer science data mining semi-supervised learning classification data stream. |
title | Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification |
title_full | Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification |
title_fullStr | Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification |
title_full_unstemmed | Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification |
title_short | Hybrid of Active Learning and Dynamic Self-Training for Data Stream Classification |
title_sort | hybrid of active learning and dynamic self training for data stream classification |
topic | computer science data mining semi-supervised learning classification data stream. |
url | http://ijict.itrc.ac.ir/article-1-26-en.html |
work_keys_str_mv | AT mohammadrezakeyvanpour hybridofactivelearninganddynamicselftrainingfordatastreamclassification AT mahnooshkholghi hybridofactivelearninganddynamicselftrainingfordatastreamclassification AT sogolhaghani hybridofactivelearninganddynamicselftrainingfordatastreamclassification |