A Comparison of K-Means and Agglomerative Clustering for Users Segmentation based on Question Answerer Reputation in Brainly Platform

Abstract

Brainly is a question and answer (QA) site that students can use as a media for questions and answers. Students can also use Brainly to find and share educational information that helps students solve their homework problems. In Brainly, users can answer questions according to their interests. However, it could be that the interest is not necessarily following the competencies possessed. It causes many answers to the questions given not to have a high rating because the answers given are of low quality to be prioritized as the main answer. This study aims to apply the K-Means and Agglomerative Clustering methods to segment users based on the reputation of the answerers by conducting clustering based on track records in answering questions on mathematics subjects. This study used the number of the brightest scores and the number of answers that did not get a rating as the basic features for clustering. The comparison between the two methods used is based on the Silhouette Score, representing the quality of the clustering results, calculated by applying the Silhouette Coefficient method. This study result indicates that the K-Means method gives better results than the Agglomerative Clustering. The Silhouette Score generated by the K-Means method is higher at 0.9081 than the Agglomerative Clustering method, which is 0.8990, which produces two clusters or two segments.