Sentiment analysis with modality processing

Abstract

Online retailers selling goods have desired to know what the customers think about their products. Sentiment analysis has enabled the people to do so by allowing them to rate their products as positive or negative. In this paper, we present a novel method for sentiment analysis of sentences with modalities which helps to determine the polarity of a phrase that describes opposite opinions. The model determines the sentiment orientation of the sentence and the type of modality observed in the sentence. This model was implemented on customer reviews of four different products used by them and the result shows that 72% of documents were correctly classified.