Mi DSpace
Usuario
Contraseña
Please use this identifier to cite or link to this item: http://hdl.handle.net/UCSP/15768
Title: Deep neural network approaches for Spanish sentiment analysis of short texts
Authors: Ochoa Luna, José
Ari, Disraeli
Keywords: Data mining;Recurrent neural networks;Sentiment analysis;Social networking (online);Benchmark datasets;Convolutional neural network;Data augmentation;Learning approach;Neural networks model;Sentence classifications;Twitter sentences;Word representations;Deep neural networks
Issue Date: 2018
Publisher: Springer Verlag
metadata.dc.relation.uri: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85057122520&doi=10.1007%2f978-3-030-03928-8_35&partnerID=40&md5=db7e538bea67ed5945193b24af816b0b
Abstract: Sentiment Analysis has been extensively researched in the last years. While important theoretical and practical results have been obtained, there is still room for improvement. In particular, when short sentences and low resources languages are considered. Thus, in this work we focus on sentiment analysis for Spanish Twitter messages. We explore the combination of several word representations (Word2Vec, Glove, Fastext) and Deep Neural Networks models in order to classify short texts. Previous Deep Learning approaches were unable to obtain optimal results for Spanish Twitter sentence classification. Conversely, we show promising results in that direction. Our best setting combines data augmentation, three word embeddings representations, Convolutional Neural Networks and Recurrent Neural Networks. This setup allows us to obtain state-of-the-art results on the TASS/SEPLN Spanish benchmark dataset, in terms of accuracy. © Springer Nature Switzerland AG 2018.
URI: http://repositorio.ucsp.edu.pe/handle/UCSP/15768
ISBN: urn:isbn:9783030039271
ISSN: 3029743
Appears in Collections:Artículos - Ciencia de la computación

Files in This Item:
There are no files associated with this item.


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.