KOP: An opinion mining system in Konkani

Opinion Mining is an area in Natural Language Processing that is becoming increasingly popular due to its ability to reduce the time and effort involved in manually inspecting and determining the opinion of a text. However, most of the research work carried out in opinion mining predominantly focuses on the English language. In this paper, a novel approach to opinion mining in the Konkani language is proposed. The proposed model which is called KOP is an opinion mining system for texts written in Devnagiri scripted Konkani. It takes into consideration several important aspects in the language text such as negations, discourses, sarcasm and positional features, which influence the opinion being conveyed by the document text. It finally outputs the polarity of the document text as positive, negative or neutral.