Authors: Sukhnandan Kaur; Rajni Mohana
Addresses: Jaypee University of Information Technology, Waknaghat, P.O. Waknaghat, Teh Kandaghat, Distt. Solan, 173 234, (H.P.), India ' Jaypee University of Information Technology, Waknaghat, P.O. Waknaghat, Teh Kandaghat, Distt. Solan, 173 234, (H.P.), India
Abstract: The exponential growth of data on the websphere accelerated the need of extracting meaningful information from it. This information can be used for better decision making. The automatic generation of sentiments from the text is called as sentiment analysis (SA). It is a collaborative process of natural language processing and data mining. This paper tries to deeply analyse the existing research work in the area of SA. It presents work done till date and segregates it in terms of level of granularity. An ideal sentiment analyser should have the intellectual capability similar to a human being. This paper mentions a roadmap of the research directions to achieve the goal of ideal sentiment analyser. These research directions include SA based on temporal summarisation, multi-linguality, etc. This paper also mentions various research aspects to work on these research directions. Future research in this direction will also refine the performance of decision making in decision support system.
Keywords: sentiment analysis; opinion mining; knowledge base; multi-linguality; fake review detection; temporal tagging; natural language processing; NLP; data mining; research directions; decision making; decision support system; DSS; fake reviews.
International Journal of Knowledge and Learning, 2015 Vol.10 No.3, pp.296 - 323
Received: 13 Feb 2015
Accepted: 13 Aug 2015
Published online: 09 Dec 2015 *