Utomo Pujianto, Agusta Rakhmat Taufani, Luis Devvi Ratna Kus Anggraini, Deni Sutaji
Youtube is a social media that is currently widely used by the public to access and share information in the form videos. In Indonesia, video content with the theme of entertainment on youtube is more popular than video content with an educational theme. So that not many content creators make educational videos with educational themes. The data used in this study consisted of video duration, type of content, availability of subtitles, number of views, likes, and comments. This data will be used to classify the popularity of videos on YouTube using the Bagging and Boosting algorithm. The stages used in this study consisted of collecting datasets, preprocessing, clustering, sampling, implementing Bagging and Boosting algorithms. This research resulted in a fairly good performance. The best performance is obtained from implementing the decision tree algorithm with boosting techniques on the dataset carried out by the SMOTE process with an accuracy value of 81.22%, a precision of 84.26%, a recall of 81.22%, and an F-Measure of 0.8272. This research is expected to provide information to content creators about content with educational themes that are preferred by the Indonesian people. © 2021 IEEE.
Universitas Negeri Malang, Electrical Engineering Department, Malang, Indonesia; Gazi Üniversitesi, Bilgisayar Bilimleri (Computer Science), Ankara, Turkey
