Clustering student Instagram accounts using author-topic model

This study proposes topic model to cluster a group of high school teenager's Instagram account in Surabaya, Indonesia by using the author-topic models method. We collect valid 235 accounts (133 female, 102 male students). gather total 3,346 captions posts from 18 senior schools. find topics that define their Instagram's post or caption; these seven are namely: feeling, Surabaya events, photography, artists, vacation, religion and music. Through process, lowest perplexity come 90 iterations, which suggests six groups topics. The concluded based on value labelled according words included topic. photography discussed Photography, artists vacation three schools, while feeling music being two one respectively.