An Implementation of Sentiment Analysis Model for Marathi Tweets using Natural Language Processing

  • Renuka Ashokrao Naukarkar, Dr. A. N. Thakare


Although a significant amount of work in sentiment analysis has been done in the English and Hindi languages, which are the most widely spoken languages in India, there is a need to perform sentiment analysis in the Marathi language for the state of Maharashtra, which is the most populous state in India. In natural language processing, sentiment analysis is the problem of determining the orientation of opinion in a piece of text with relation to the topic under consideration. It is concerned with determining the emotions, feelings, and attitude of a speaker or writer from a given piece of written material. Sentiment Analysis is the process of extracting information from text about a user's behaviour, likes, and dislikes of that individual. The goal of SA is to find opinions, determine the sentiments expressed by those opinions, and then classify them according to their polarity. It is proposed that a Gate Processor (natural language processor) be used to determine the overall polarity of the document; the final aggregated polarity can be either positive, negative, or neutral. The objective was to develop an accurate methodology for evaluating thoughts and opinions from user generated content in Marathi, with a particular emphasis on investigating the influence of negative sentiment.