SENTIMENT ANALYSIS OF MARATHI LANGUAGE

Autor: SUJATA DESHMUKH, NILEEMA PATIL, SURABHI ROTIWAR, JASON NUNES
Jazyk: angličtina
Rok vydání: 2017
Předmět:
DOI: 10.5281/zenodo.1445362
Popis: Sentiment analysis offers many benefits and opportunities from business, government and consumer perspective in this digital data explosive age. According to Google, in partnership with KPMG India report (April-2017), titled Indian Languages Defining Indias Internet currently India today has 234 million Indian Language users who are online, compared to 175 million English web users and expected 536 million Indians to use regional languages while online by 2021. However Marathi users are expected to make significant contribution to define India Internet volume. By considering aspects of Marathi language and benefits of sentiment analysis, this paper presents a approach to overcome the barriers and difficulties being faced for analyzing text in Marathi language. The proposed system detects hidden sentiments in text of Marathi language. The system uses sentiment analysis methodology in order to achieve desired functionality. In this system, a corpus based approach is proposed, i.e the creation of a diverse up to date corpus of Marathi keywords along with their individual polarities with respect to the Word Net which is consider as a corpus. The algorithm is used to calculate the cumulative polarity of the text and rank the sentence as positive negative or neutral on a set scale standard. https://journalnx.com/journal-article/20150361
Databáze: OpenAIRE