Exploiting Propositions for Opinion Mining

摘要: With different social media and commercial platforms, users express their opinion about products in a textual form. Automatically extracting the polarity (i.e. whether is positive or negative) of user can be useful for both actors: online platform incorporating feedback to improve product as well client who might get recommendations according his her preferences. Different approaches tackling problem, have been suggested mainly using syntactic features. The “Challenge on Semantic Sentiment Analysis” aims go beyond word-level analysis by semantic information. In this paper we propose novel approach employing information grammatical unit called preposition. We try drive target review from summary information, which serves an input identify proposition it. Our implementation relies hypothesis that expressing summary, usually containing main