Description

About The Book

The aim of this book is to consign the issue of sentiment analysis in twitter, which means classifying tweets in accordance with the sentiment expressed in them by a user or an organization: positive, negative or neutral. Twitter is an online micro-blogging and social-networking stage which lets users/handlers to write small status updates. It is a hastily escalating facility with over 330 million monthly active users (as per the report available of 2019) – out of which 130 million are active users and more than half of them log on twitter on a day-to-day basis – producing of about 350 million tweets per day. Due to this huge amount of usage we hope to attain a replication of community sentiment by analyzing the sentiments articulated in the tweets. Investigating the public sentiment is imperative for many applications such as organizations trying to find out the response of their products in the market, predicting political appointments and predicting socioeconomic phenomena like stock market.
In this book, we examine the effectiveness of dialectal features for detecting the sentiment of Twitter messages. We measure the utility of existing lexical properties as well as features that capture data about the formal, informal, and creative language used in microblogging platforms such as Twitter.
Sentiment analysis over Twitter suggest governments, administrations and other companies a fast and functioning way to monitor the people’s state of mind towards their brand, corporate, directors, etc. An extensive variety of features and procedures for training sentiment classifiers for Twitter datasets have been examined in recent years with fluctuating results. In our project, we presented a novel method of adding semantics as added structures into the training set for sentiment analysis. For each extracted entity (e.g. iPhone) from tweets, we add its semantic concept as an additional feature and measure the correlation of the representative concept with negative/positive sentiment.