Improved Accuracy of Naïve Bayes Algorithm and Support Vector Machine Using Particle Swarm Optimization for Menstrual Cup Sentiment Analysis on Twitter

Abstract

Menstrual cup is a menstrual hygiene sanitation tool that replaces disposable sanitary napkins for women that reaps many pros and cons in its use. From this, it is necessary to analyze the public's views regarding the use of menstrual cups, which is called sentiment analysis. Sentiment analysis is a process that aims to determine the polarity of the sentiment of a text. This paper performs a classification of menstrual cup sentiment analysis on Twitter using the Naïve Bayes and the Support Vector Machine algorithm. Particle Swarm Optimization is applied to improve the accuracy of both classification algorithms. The final result of the accuracy obtained by the Naïve Bayes algorithm is 92.72% and the Support Vector Machine algorithm is 96.13%. While the accuracy results after Particle Swarm Optimization is applied, for Naïve Bayes it produces an accuracy rate of 95.87%, and Support Vector Machine is 96.68%.

Performance evaluation of classifiers for the COVID-19 symptom-based dataset using different feature selection methods

“…Filter methods, on the other hand is one of the popular feature selection methods which include chisquared [14,15], information gain [12,16], and correlation-based feature selection (CFS) [17][18][19]. The CFS algorithm, which has been shown to effectively reduce feature dimensionality while maintaining classification accuracy in medical datasets.…”