Batch size for training convolutional neural networks for sentence classification

Nabeel Zuhair Tawfeeq Abdulnabi and Oguz Altun
Additional contact information
Nabeel Zuhair Tawfeeq Abdulnabi: Computer Engineering Department, Yildiz Technical University, Istanbul, Turkey
Oguz Altun: Computer Engineering Department, Yildiz Technical University, Istanbul, Turkey

Journal of Advances in Technology and Engineering Research, 2016, vol. 2, issue 5, 156-163

Abstract: Sentence classification of shortened text such as single sentences of movie review is a hard subject because of the limited finite information that they normally contain. We present a Convolutional Neural Network (CNN) architecture and better hyper-parameter values for learning sentence classi??ication with no preprocessing on small sized data. The CNN used in this work have multiple stages. First the input layer consist of sentence concatenated word embedding. Then followed by convolutional layer with different ??ilter sizes for learning sentence level features, followed by max-pooling layer which concatenate features to form ??inal feature vector. Lastly a softmax classi??ier is used. In our work we allow network to handle arbitrarily batch size with different dropout ratios, which is gave us an excellent way to regularize our CNN and block neurons from co-adapting and impose them to learn useful features. By using CNN with multi ??ilter sizes we can detect speci??ic features such as existence of negations like “not amazing”. Our approach achieves state-of-the-art result for sentence sentiment prediction in both binary positive/negative classification.

Keywords: Convolutional Neural Network; Sentence Classi?ication; Word embedding (search for similar items in EconPapers)
Date: 2016
References: View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
https://tafpublications.com/platform/Articles/full-jater2.5.3.php (application/pdf)
https://tafpublications.com/gip_content/paper/jater-2.5.3.pdf (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:apb:jaterr:2016:p:156-163

DOI: 10.20474/jater-2.5.3

Access Statistics for this article

Journal of Advances in Technology and Engineering Research is currently edited by A/Professor Akbar A. Khatibi

More articles in Journal of Advances in Technology and Engineering Research from A/Professor Akbar A. Khatibi Calle Alarcon 66, Sant Adrian De Besos 08930, Barcelona Spain.
Bibliographic data for series maintained by A/Professor Akbar A. Khatibi ().