Sentiment Analysis on Brazilian Portuguese User Reviews

TL;DR

This paper evaluates various document embedding strategies for sentiment analysis on Brazilian Portuguese user reviews, providing a unified dataset and benchmark to assess model generalization across different datasets.

Contribution

It introduces a unified dataset with standardized partitions for Brazilian Portuguese sentiment analysis and evaluates model generalization across multiple datasets.

Findings

Document embedding strategies vary in effectiveness.

Unified dataset enables fair comparison of models.

Models show limited generalization across datasets.

Abstract

Sentiment Analysis is one of the most classical and primarily studied natural language processing tasks. This problem had a notable advance with the proposition of more complex and scalable machine learning models. Despite this progress, the Brazilian Portuguese language still disposes only of limited linguistic resources, such as datasets dedicated to sentiment classification, especially when considering the existence of predefined partitions in training, testing, and validation sets that would allow a more fair comparison of different algorithm alternatives. Motivated by these issues, this work analyzes the predictive performance of a range of document embedding strategies, assuming the polarity as the system outcome. This analysis includes five sentiment analysis datasets in Brazilian Portuguese, unified in a single dataset, and a reference partitioning in training, testing, and validation sets, both made publicly available through a digital repository. A cross-evaluation of dataset-specific models over different contexts is conducted to evaluate their generalization capabilities and the feasibility of adopting a unique model for addressing all scenarios.