Machine Learning Based Sentiment Analysis on Twitter Data

Abstract

This paper addresses the problem of sentiment analysis in twitter; that is characterizing tweets as per the feeling communicated in them: positive, negative, or impartial. Twitter is online small-scale blogging and person to person communication stage which permits clients to compose short notices of greatest length 140 characters. It is a quickly extending administration with more than 200 million enlisted clients-out of which 100 million are dynamic clients and half of them sign on twitter day by day-creating almost 250 million tweets for every day. Because of this huge measure of use, we would like to accomplish an impression of open estimation by dissecting the notions communicated in the tweets. Tearing down the open evaluation is critical for certain applications, for instance, firms endeavouring to find the response of their things in the market, predicting political choices and envisioning monetary wonders like the stock exchange. The point of this task is to build up a practical classifier for exact and programmed feeling grouping of an obscure tweet stream by utilizing the convolutional neural system. © 2020, World Academy of Research in Science and Engineering. All rights reserved.