Please use this identifier to cite or link to this item:
http://hdl.handle.net/10071/23676| Author(s): | Tavares, C. Ribeiro, R. Batista, F. |
| Editor: | Queirós. R., Pinto, M., Simões, A., Portela, F., & Pereira, M. J. |
| Date: | 2021 |
| Title: | Sentiment analysis of Portuguese economic news |
| Volume: | 94 |
| Event title: | 10th Symposium on Languages, Applications and Technologies, SLATE 2021 |
| ISSN: | 2190-6807 |
| ISBN: | 978-3-95977-202-0 |
| DOI (Digital Object Identifier): | 10.4230/OASIcs.SLATE.2021.17 |
| Keywords: | Sentiment analysis Economic news Portuguese language |
| Abstract: | This paper proposes a rule-based method for automatic polarity detection over economic news texts, which proved suitable for detecting the sentiment in Portuguese economic news. The data used in our experiments consists of 400 manually annotated sentences extracted from economic news, used for evaluation, and about 90 thousand Portuguese economic news, extracted from two well-known Portuguese newspapers, covering the period from 2010 to 2020, that have been used for training our systems. In order to perform sentiment analysis of economic news, we have also tested the adaptation of existing pre-trained modules, and also performed experiments with a set of Machine Learning approaches, and self-training. Experimental results show that our rule-based approach, that uses manually written rules related to the economic context, achieves the best results for automatically detecting the polarity of economic news, largely surpassing the other approaches. |
| Peerreviewed: | yes |
| Access type: | Open Access |
| Appears in Collections: | CTI-CRI - Comunicações a conferências internacionais |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| conferenceobject_82720.pdf | Versão Editora | 550,02 kB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.