Abstract

Sentiment analysis functions by analyzing and extracting opinions from documents, websites, blogs, discussion forums and others to identify sentiment patterns on opinions expressed by consumers. It analyzes people's sentiment and identifies types of sentiment in comments expressed by consumers on certain matters. This paper highlights comparative studies on the types of feature selection in sentiment analysis based on natural language processing and modern methods such as Genetic Algorithm and Rough Set Theory. This study compares feature selection in text classification based on traditional and sentiment analysis methods. Feature selection is an important step in sentiment analysis because a suitable feature selection can identify the actual product features criticized or discussed by consumers. It can be concluded that metaheuristic based algorithms have the potential to be implemented in sentiment analysis research and can produce an optimal subset of features by eliminating features that are irrelevant and redundant.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call