SMART TRASH MENGGUNAKAN METODE CLUSTERING DENGAN PENDEKATAN CENTROID LINKAGE

Abstract

The focus of the research study is related to automatic waste handling with a microcontroller-based Smart Trash System. The purpose of this study is the implementation of the clustering method with the centroid linkage approach on Smart Trash. The term Smart Trash is intended for bins that have features with more functions than bins in general. This Smart Trash System Design uses Arduino UNO as a data processor, inductive proximity sensor and proximity sensor as a metal and non-metal waste detector. This research uses clustering method with centroid linkage approach. The results of inductive sensor metal waste detection devices using clustering method with the centroid linkage approach can detect waste 0-4 mm away from garbage objects with an accuracy value of 71% and, the ultrasonic sensor non-metal sensor waste detection devices use the clustering method with the centroid linkage approach can detect rubbish with a distance of 3-18 cm from the garbage object with an accuracy value of 85.7%.