Sentiment Analysis for Product and Organization Reviews using RobERTa Model

Social media related to healthcare, crime, finance, academy, military, and travel uses natural language processing to determine and classify expressions mentioned in the text format through sentiment analysis. Reviews about goods or other realities, such as political leaders, government programs, etc., are what we refer to as concept-position sentiment analysis. Previous work on conception-position sentiment analysis has employed a small number of verbal rules for rooting generalities and their features. In addition, it demonstrates several ways to perform sentiment analysis and effective methodology. In this paper, the Roberta model has been used for sentiment analysis and compared with VADER. Using the RobERTa model, the accuracy of sentiment analysis is more sustainable to improve the efficiency of our proposed system. In comparison with the BERT model or VADER, we achieved 88.5% accuracy in sentiment analysis.