Please use this identifier to cite or link to this item:
|Title: ||Handwritten Character Recognition Using Active Semi-supervised Learning|
|Authors: ||Inkeaw, Papangkorn|
|Editors: ||Yin, Hujun|
Tallón-Ballesteros, Antonio J.
|Keywords: ||Handwritten character recognition|
|Issue Date: ||2018|
|Citation: ||Papangkorn Inkeaw, Jakramate Bootkrajang, Teresa Gonçalves, and Jeerayut Chaija-
ruwanich. Handwritten Character Recognition Using Active Semi-supervised Learning.
In Hujun Yin, David Camacho, Paulo Novais, and Antonio J. Tallón-Ballesteros, editors,
Intelligent Data Engineering and Automated Learning, IDEAL 2018, volume 11314 of
Lecture Notes in Computer Science, pages 69–78, Cham, 2018. Springer. ISBN 978-3-
|Abstract: ||Constructing a handwritten character recognition model is considered challenging partly due to the high variety of handwriting
styles and the limited amount of training data. In practice, only a handful of labeled examples from limited number of writers are provided during the training of the model. Still, a large collection of already available unlabeled handwritten character data from several sources are often left
unused. To alleviate the problem of small training sample size, we propose a graph-based active semi-supervised learning approach for handwritten character recognizer construction. The method iteratively builds a
neighborhood graph of all examples including the unlabeled ones, assigns pseudo labels to the unlabeled data and retrains the model. Additionally, the label of the least confident pseudo label according to a newly
proposed uncertainty measure is to be requested from the oracle. Experiments on NIST handwritten digits dataset demonstrated that the proposed learning method better utilizes the unlabeled data compared to
existing approaches as measured by recognition accuracy. In addition, our active learning strategy is also more effective compared to baseline strategies.|
|Appears in Collections:||INF - Artigos em Livros de Actas/Proceedings|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.