Evaluating retraining rules for semi-supervised learning in neural network based cursive word recognition


Autoria(s): Frinken, Volkmar; Bunke, Horst
Data(s)

2009

Resumo

Training a system to recognize handwritten words is a task that requires a large amount of data with their correct transcription. However, the creation of such a training set, including the generation of the ground truth, is tedious and costly. One way of reducing the high cost of labeled training data acquisition is to exploit unlabeled data, which can be gathered easily. Making use of both labeled and unlabeled data is known as semi-supervised learning. One of the most general versions of semi-supervised learning is self-training, where a recognizer iteratively retrains itself on its own output on new, unlabeled data. In this paper we propose to apply semi-supervised learning, and in particular self-training, to the problem of cursive, handwritten word recognition. The special focus of the paper is on retraining rules that define what data are actually being used in the retraining phase. In a series of experiments it is shown that the performance of a neural network based recognizer can be significantly improved through the use of unlabeled data and self-training if appropriate retraining rules are applied.

Formato

application/pdf

Identificador

http://boris.unibe.ch/37085/1/05277801.pdf

Frinken, Volkmar; Bunke, Horst (2009). Evaluating retraining rules for semi-supervised learning in neural network based cursive word recognition. In: 10th International Conference on Document Analysis and Recognition ICDAR 2009 (pp. 31-35). Washington, DC: IEEE Computer Society 10.1109/ICDAR.2009.18 <http://dx.doi.org/10.1109/ICDAR.2009.18>

doi:10.7892/boris.37085

info:doi:10.1109/ICDAR.2009.18

urn:issn:1520-5363

urn:isbn:978-0-7695-3725-2

Idioma(s)

eng

Publicador

IEEE Computer Society

Relação

http://boris.unibe.ch/37085/

Direitos

info:eu-repo/semantics/openAccess

Fonte

Frinken, Volkmar; Bunke, Horst (2009). Evaluating retraining rules for semi-supervised learning in neural network based cursive word recognition. In: 10th International Conference on Document Analysis and Recognition ICDAR 2009 (pp. 31-35). Washington, DC: IEEE Computer Society 10.1109/ICDAR.2009.18 <http://dx.doi.org/10.1109/ICDAR.2009.18>

Palavras-Chave #000 Computer science, knowledge & systems #510 Mathematics
Tipo

info:eu-repo/semantics/conferenceObject

info:eu-repo/semantics/publishedVersion

PeerReviewed