Author : Miss.Priyanka Pawar and Prof. Dr.Pankaj Agarkar

Sentiment analysis alludes to the application for processing natural language, content investigation, computational etymology to deliberately perceive, remove, evaluate, and learn full of feeling states and abstract data. Twitter, being one among a few well known web based life stages, is where individuals frequently decide to express their feelings and notions about a brand, an item or a help. Recent studies show that patterns of spreading feelings on Twitter have close relationships with the polarities of Twitter messages. In this paper center around how to consolidate the literary data of Twitter messages and notion spread models to show signs of improvement execution of slant examination in Twitter information. To this end, proposed framework first analyses the dissemination of feelings by considering a wonder considered reversal of emotions and locate some fascinating properties of the inversion of emotions. Therefore we consider the interrelations between the textual information of Twitter messages and the patterns of diffusion of feelings, and propose random forest machine learning to predict the polarities of the feelings expressed in Twitter messages. Apparently, this work is the first to utilize estimation dispersal models to improve Twitter’s notion examination. Various tests in reality dataset show that, contrasted with cutting edge content based examination calculations.

Keywords: Text Mining,  Machine learning, Supervised,Sentiment analysis, sentiment diffusion, Twitter