A Mathematical Prediction Model on Sentiment analysis of Twitter data

Sentiment analysis is a process of analyzing textual data to extract sentiment or feeling of textual data on social networks shares by the millions of online people or user’s in the Internet world. The current scenario of research in the pasture of sentiment analysis is focused on social network platforms like Twitter, Facebook, etc. The sentiment of data is based on the textual data on social networks. Sentiment analysis is a procedure to extract the positive, negative or neutral sentiment in the form of numeric values range from -1 to 1. It is a sentiment of a user's expression in the form of text about any people or product review, etc. The sentiment of textual data is based on the feature extraction of a text or content Most of the researchers working on sentiment analysis focus on varieties of features of tweets. In feature extraction, we focused on the combined length of token and character in the textual data. In this paper, we had collected approximately three thousand two hundred tweets from Twitter and extract the sentiment of tweets and count the number of characters or the length of tweets using machine learning concepts in python platforms. We proposed a mathematical model to predict a relationship among the lengths of textual data or the number of characters in tweets of users and its corresponding sentiments using machine learning and linear regression.