Twitter Data Mining Model for Social Happiness by Sentiment Analysis Using Distant Supervision and Naïve Bayes Learning Algorithm

Authors

  • Prasetyo Nugraha Gema, Suharjito

Abstract

Social happiness is one of the most important aspects in the century. More
institutions, NGOs and researchers start to devise methods which can accurately
and efficiently measure it. The primary idea underlying this research is that
happiness is a collection of sentiments; and sentiments are traceable through social
media posts. Happiness is modelled using words with thehelp of machine learning
techniques such as Naïve Bayes and Distant Supervision. This research aims to
find out how well automated sentiment analysisusing machine learning techniques
is able to work hand in hand. The result shows that the method achieves great
consistency of 0.5468 chi square score (required less than 3.325) and sensitivity:
2,075.288 chi square score (required more than 1.675). Keywords—Social
Happiness; Data Mining; Twitter; Naïve Bayes; Distant Supervision;

Published

2020-06-30

Issue

Section

Articles