PENERAPAN ALGORITMA NAIVE BAYES DAN PARTICLE SWARM OPTIMIZATION UNTUK KLASIFIKASI BERITA HOAX PADA MEDIA SOSIAL

APPLICATION OF NAIVE BAYES ALGORITHM AND PARTICLE SWARM OPTIMIZATION FOR CLASSIFICATION OF HOAX NEWS IN SOCIAL MEDIA

Abstract

Social media is the most effective way to facilitate fast information, unfortunately, there are some elements who use social media to add hoax or deception to give misleading opinions to the public. Therefore a method is needed to classify hoax news and non-hoax news on social media. Naive Bayes is a simple classification algorithm but has high qualifications, but Naive Bayes has a very sensitive shortcoming in the selection of features and therefore the Particle Swarm Optimization method is needed to improve the expected results. After conducting research with the Naive Bayes method and the Naive Bayes method based on Particle Swarm Optimization, the results obtained are Naive Bayes yielding 74.67% while the Naive Bayes based on Particle Swarm Optimization with an accuracy value of 85.19%. The purpose of this study is to see a large comparison. Swarm Optimization particles to improve accuracy in the classification of hoax news on social media using the Naive Bayes classifier. After using Particle Swarm Optimization the test results increased by 10.52%.