Implementasi Algoritma C5.0 Pada Smart Lamp Berbasis Website

Abstract

Lighting from lamps plays an important role, but manually controlling lamps often makes users forget to turn off the lights when leaving home. This research introduces smartlamp, which is a lamp integrated with IoT system. In this research, smartlamp is designed to provide automatic control by recognizing the habit data of home occupants using the C5.0 Algorithm method. This system focuses on recognizing data on the habits of home residents using the C5.0 Algorithm method with the use of 864 training data conducted for 20 days and 48 testing data for 2 days resulting in system implementation with an accuracy of 98.61%, recall 98.47%, precision 98.47%, and an error rate of 1.39%. The results of this study show the ability of the C5.0 Algorithm method to recognize user habit data. Suggestions for further research are to build a system that uses a bluetooth connection so that it can be used when there is no internet network and add several sensors to measure the level of efficiency of power usage at home