Exploring the use of BERT in Sentiment Analysis of Evaluation of Amazon Reviews

Jason Dolorso, Nika Karen Espiritu, Alejandro White
Asian Institute of Management

Abstract

Sentiment analysis problems have frequently used machine learning models and lexicon-based methods. In this paper, we explore the use of a state-of-the-art model called BERT, or Bidirectional Encoder Representations from Transformers. We conducted a comprehensive review of the literature in sentiment analysis. In this study, we compared the performance of BERT in sentiment analysis to NLTK’s VADER, TextBlob’s Sentiment Analyzer, and Logistic Regression. BERT had the highest accuracy and recall score, but also took the longest runtime.

Keywords

Sentiment analysis, VADER, BERT

---

Full text article and source codes can be provided upon request.