Website Forum Diskusi Dengan Filter Spam Menggunakan Metode Clustering K-Medoids dan Metode Stemming Tala

Abstract

In online learning, there are known terms of synchronous learning and asynchronous learning. Some common asynchronous learning activities are interacting with the Learning Management System (LMS), communicating using email, posting on discussion forums and reading articles. In this study, a discussion forum will be designed to support the online learning process. In discussion forums, there are often threads that are opened containing messages that are not useful or not related to the topic being discussed. These messages are often referred to as spam. To solve this problem, the K-Medoids clustering algorithm can be applied. Meanwhile, to carry out the stemming process, the Tala method will be used. The result of this research is a discussion forum website application that provides a spam filter feature so that discussion forums can be clean from spam. The website also supports the process of adding and storing new spam data lists, so that users can enter new spam words to improve spam detection accuracy.