Using Sentiment Analysis to Determine Users' Likes on Twitter
Yo‐Ping Huang, Nontobeko Hlongwane, Li-Jen Kao · 2018
Nowadays the undeniable truth is that social media users are more likely to share how they feel about a current "hot topic" on social media platforms. Hot topics are current affairs that may trend regionally or globally. Users may post negative, positive or neutral opinions about that topic or a particular product they are using. The advancement of artificial intelligence (AI) has opened doors in which we can write algorithms to help users detect and categorize online opinions. This study aims at proposing an AI model to detect emotions in unstructured texts. We analyze the sentiments of user views about the recently controversial issues and compare them with the related popular topics. The major contribution is to leverage on social media to estimate a sentimental opinion assessment on the latest trends or topics of controversy on Twitter. Our main objective is to obtain perceptions of the users' opinions based on the number of likes, retweets and using a natural language processing (NLP) toolkit to determine the sentiment of texts. Experimental results confirm that sentiment analysis is valuable to identify users' likes, comments, and retweets on a product.