Sentiment Analysis at the Product Feature Level and Based on Users Gender

Abstract

Social media data is one of the most effective and accurate indicators of public sentiment, so that analyzing this information can provide researchers with interesting results from users' sentiment about characters, subjects, products, and services. In this study, while reviewing users' opinion on Twitter about the various features of two competing mobile phone products on the market, Apple's Iphone X and Samsung's Galaxy S9, we examine their sentiment based on the gender of consumers of these two products. This study is performed using the relation-based method in the feature extraction step and Lexicon-Based in the polarity of opinions step. The results of this study show that the popularity of different product features varies between male and female users, and based on these results, business owners can produce products that focus on people's gender or design smart advertising plan according to their interests. These measures ultimately lead to increased business profitability and customer satisfaction.

Keywords

• sentiment analysis
• twitter
• feature extraction
• polarity of opinions