Capturing people interests and emotions through natural language and fuzzy linguistic terms

"The main objective of this project is to define a methodology able to find central sentiment and degree of consensus among a set of reviews in a social network context or in news context. In this project, a hybrid methodology is proposed. On the one hand, we will apply content and sentiment analysis on Big Data extracted from media or social media to identify the aspects or topics with positive, negative or neutral sentiment opinions. On the other hand, a methodology based on fuzzy linguistic descriptions will be used to capture from these opinions a fuzzy description of their concerns and emotions and their degree of consensus or agreement on specific topics.
In this framework, sentiment will be extracted from text analysis and then will be expressed in fuzzy linguistic terms. The centralized sentiment will be defined to reflect the specific opinion on a certain domain and a comparison will be made among the positions taken by other users or in other contexts. "