Playing with Word Meanings

With the wide-world expansion of the social web, subjectivity analysis became lately one of the main research focus in the area of intelligent information retrieval. Being able to find out what people feel about a specific topic, be it a marketed product, a public person or a political issue, represents a very interesting application for a large class of actors, from the everyday product and service consumers, to the marketing and political analysts or decision-makers. In this paper we will diverge from the standard goal of subjectivity analysis i.e. computing the polarity of an opinionated sentence. Instead, we will try to detect those sentences of a given text that might have different connotations when put in different contexts. We describe a system called CONAN, initially designed to be a subjectivity analyzer, but recently extended towards detecting possible connotations shifts.