Machine Learning Approaches to Sentiment Analysis in Social Networks Using Political Tweets

Abstract

Online social networks offer a quantitative
assessment of people's psychological behavior and aid
in the general analysis of social or political concerns. In
text mining research, opinions, attitudes, and subjectivity
in text and other expressions are ascertained by a
computational method. Furthermore, the majority of
approaches try to simulate word syntactic information
without taking sentiment into account. A brief description
of the various machine learning (ML) models utilized in
sentiment analysis is provided in this paper. Additionally,
suggest a productive modular strategy that will provide
exact correctness when testing and evaluating the
Twitter data. In today's world, when national and
international leaders are important, political reviews
linked to Twitter data collection are more prevalent. Our
work's goal is to find solutions by analyzing and
contrasting various approaches. According to a
simulation study, there is a practical approach to
comprehensively analyze and use a political twitter
dataset regarding an international leader, while
concentrating on additional sentiment dataset validation
to increase the precision of tweet sentiment analysis.