Plant Recomendation Using Triple Exponential Smoothing and K-Nearest Neighbor Based on Internet of Things
Muhammad Jurnalies Habbibie (a*), Yaddarabullah (a), Ade Syahputra (a)

a) Program Studi Teknik Informatika, Universitas Trilogi
Jl. TMP Kalibata No. 1, Jakarta 12760, Indonesia
* mjurnaliesh[at]trilogi.ac.id

Abstract

Nutritious Garden Trilogy University is an agricultural land managed by urban farmers as well as students who are also lecturers to practice teaching and learn in the cultivation of types of plant commodities. Cultivation of plant species done by examining the climatic factors of plants on the land. The sustainability reason is one of the obstacles to make sure the results of cultivation in the teaching and learning process. The changing climate makes urban farmers get trouble to determine types of plants to be planted. This study will develop a system to forgive the recommendation of the type of plant according to the change of climate. The climate changes recorded using the internet of things sensors which consist of temperature, humidity, light intensity, and wind speed. The data entered will be processed using the triple exponential smoothing method as a forecast to predict future weather, then classified using the k-nearest neighbor to get the types of plants. The results of forecasting testing from sensors using the mean absolute percentage error obtained values of 9.53% temperature, 16.44% humidity, 3.73% light intensity, 19.42% wind speed.

Keywords: Plant Commodity, Internet of Things, Triple Exponential Smoothing, K-Nearest Neighbor

Topic: Information Engineering