Fuzzy sentiment classification in social network Facebook' statuses mining

Nowadays, posts expressing opinions in social networks are beneficial for businesses, sway public sentiments and emotions having higher social and political impact. In fact, opinions mining or sentiment classification is an important issue in social networks. Therefore, machine learning methods and natural language processing are very effective for opinion mining which are widely used in social media network. The main purpose of the paper is to demonstrate the feasibility of machine learning and sentiment analysis for studying Tunisian users' statuses. We aim to extract terms related to sentiment and behavior, especially during the “Tunisian Election”. While existing sentiment analysis methods focus only on the extraction of positive and negative opinions, in our work we aim to extract fuzzy opinions. That is why we propose to follow Fuzzy Support Vector Machine during our classification process. This later is compared with the basic Support Vector Machine referring to assessement measures such us accuracy, precision, recall, and F-measure.