Perbandingan Algoritma Naive Bayes Classifier dengan K-Nearest Neighbor (K-NN) Pada Ulasan Aplikasi Youtube Di PlayStore

Azmi Ulwan Dasuki
Erizal

Prosiding Seminar Nasional Teknoka
Vol 9 (2024): Proceeding of TEKNOKA National Seminar - 9

Youtube is an application that is still widely used today where Youtube users can watch and post various types of videos ranging from entertainment, news, and insights for free. With the large number of Youtube application users, more and more people are giving comments ranging from positive to negative comments on the Youtube application on the Google Play Store. The purpose of this study is to analyze the sentiment of Youtube application users based on reviews on the Google Play Store. This study went through the stages of data extraction, data labeling, text preprocessing, and also the classification of the Naive Bayes Classifier and K-NN algorithms. After going through the Naive Bayes Classifier and K-NN classification stages, namely the K-NN algorithm produced 81,26% with a precision value of 83,29% and a recall value of 96,29% while the Naive Bayes Classifier algorithm only produced an accuracy value of 78,24%, precision of 82,83% and recall of 92,43%. This means that from the results of the sentiment analysis, the K-NN algorithm is superior to the Naive Bayes Classifier algorithm and public opinion towards the YouTube application is more negative than positive.