Analysis and design of intelligent approaches for relationship extraction and sentiment analysis

Abstract

Predicting future events has always been an interesting task from predicting weather and natural disaster to predicting sports outcomes election results disease outbreak and stock market assets It seems that it is in the human nature to try to guess or calculate what will happen next Moreover with the advancement of computer science and methodologies for data analysis predicting future trends and events has become easier and more accurate newlineMotivated by these phenomena and earlier studies, this thesis investigates whether the opinions expressed through social media platforms can be used for extracting the relationships and sentiments of public on multiple events To detect expressed opinions on social media a research has been performed for detecting opinions relationships and emotions in texts newlineEvery major event in the world has an online presence and social media users use social media platforms to express their sentiments and opinions towards it The thesis first presents the experiments which resulted in selecting the most suitable sentiment analysis algorithm and determining the best data preprocessing methodology newlineUse of social media platforms such as Twitter encompasses an edge of advantage over another medium due to its direct access to information about events easily accessible at nearly zero cost real time analysis and automated language processing capabilities An effective multi dimensional prediction framework using Twitter has been developed The prediction model is developed using dynamic keywords from RSS feeds with trending keywords from Twitter newlineExtraction of emotions from text posted on social media by different categories of users is one of the crucial tasks in the field of opining mining and sentiment analysis An advanced framework for automatic detection of emotions of users in Multilanguage text data using emotion theories which deal with linguistics and psychology has been developed newlineSocial media data has been used for the domain of medical sciences for improving the facilities and enhancement