Please use this identifier to cite or link to this item: https://doi.org/10.21256/zhaw-1526
Title: Potential and limitations of cross-domain sentiment classification
Authors : von Grünigen, Dirk
Weilenmann, Martin
Deriu, Jan Milan
Cieliebak, Mark
Proceedings: Proceedings of the Fifth International Workshop on Natural Language Processing for Social Media
Pages : 17
Pages to: 24
Conference details: Fifth International Workshop on Natural Language Processing for Social Media, Valencia, Spain, April 3-7, 2017
Publisher / Ed. Institution : Association for Computational Linguistics
Publisher / Ed. Institution: Stroudsburg PA, USA
Issue Date: Apr-2017
Language : Englisch / English
Subjects : Sentiment Analysis
Subject (DDC) : 004: Informatik
005: Computerprogramme, Datenverarbeitung
Abstract: In this paper we investigate the cross-domain performance of sentiment analysis systems. For this purpose we train a con-volutional neural network (CNN) on data from different domains and evaluate its performance on other domains. Furthermore , we evaluate the usefulness of combining a large amount of different smaller annotated corpora to a large corpus. Our results show that more sophisticated approaches are required to train a system that works equally well on various domains.
Departement: School of Engineering
Organisational Unit: Institut für Angewandte Informationstechnologie (InIT)
Publication type: Konferenz: Paper / Conference Paper
DOI : 10.18653/v1/W17-1103
10.21256/zhaw-1526
ISBN: 9781510838710
URI: https://digitalcollection.zhaw.ch/handle/11475/1852
Appears in Collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
W17-1103.pdf1.67 MBAdobe PDFThumbnail
View/Open


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