An improved active learning method based on feature selection

DOI

10.2991/asei-15.2015.37How to use a DOI?

Keywords

active learning, support vector machine, principal component analysis, PCA

Abstract

An improved active learning method taking advantage of feature selection technique is proposed. In early stages of active learning, the whole dataset is described using only the few key features, so that its overall distribution characteristic can be learned easily, reducing active learning’s possibility of falling into bad local optimum. As active learning proceeds, more and more data get labeled. Only then are detailed features of the dataset gradually added to further enhance the model’s classification performance. Experiments show that it is more efficient and more robust than traditional technique.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chunjiang Fu
AU  - Liang Gong
AU  - Yupu Yang
PY  - 2015/05
DA  - 2015/05
TI  - An improved active learning method based on feature selection
BT  - Proceedings of the 2015 International conference on Applied Science and Engineering Innovation
PB  - Atlantis Press
SP  - 170
EP  - 174
SN  - 2352-5401
UR  - https://doi.org/10.2991/asei-15.2015.37
DO  - 10.2991/asei-15.2015.37
ID  - Fu2015/05
ER  -