Exploring BERT for Aspect-based Sentiment Analysis in Portuguese Language

作者

  • Émerson Philippe Lopes Federal University of Pelotas - UFPel
  • Larissa Freitas Federal University of Pelotas - UFPel
  • Gabriel Gomes Federal University of Pelotas - UFPel
  • Gerônimo Lemos Federal University of Pelotas - UFPel
  • Luiz Otávio Hammes Federal University of Pelotas - UFPel
  • Ulisses B. Corrêa Federal University of Pelotas - UFPel

##plugins.pubIds.doi.readerDisplayName##:

https://doi.org/10.32473/flairs.v35i.130601

关键词:

Aspect-Based Sentiment Analysis, BERT, Portuguese Language, Aspect Sentiment Classification

摘要

Aspect-Based Sentiment Analysis (ABSA) is a Natural Language Processing (NLP) task that extracts referred aspects from text and assigns polarities to opinions about those aspects. Most research on ABSA focuses on English. Only a few ABSA works deal with the Portuguese language. In this work, we used BERTimbau to create a Question-Answer approach to ABSA in Portuguese. First, we post-trained this model with text from the same domain as our target corpus. Then, we constructed an auxiliary sentence from the aspect and converted ABSA to a sentence-pair classification task, such as question answering (QA) and natural language inference (NLI). Our experiments show that ABSA based on BERT for Portuguese achieved Balanced Accuracy (BACC) of 77% on a corpus of reviews about the accommodation sector using a post-trained model with QA approach.

##submission.downloads##

已出版

2022-05-04

##submission.howToCite##

Lopes, Émerson P., Freitas, L., Gomes, G., Lemos, G., Hammes, L. O., & Corrêa, U. B. (2022). Exploring BERT for Aspect-based Sentiment Analysis in Portuguese Language. The International FLAIRS Conference Proceedings, 35. https://doi.org/10.32473/flairs.v35i.130601

期

卷 35 (2022): Proceedings of FLAIRS-35

栏目

Special Track: Applied Natural Language Processing

##submission.license##

##submission.copyrightStatement##

##submission.license.cc.by-nc4.footer##