Design of Website-Based Sugarcane Forecasting Information System
Masud Effendi *; Dyah Dwi Yuliani; Danang Triagus Setyawan; Usman Effendi
Department of Agroindustrial Technology, Faculty of Agricultural Technology, Universitas Brawijaya
Veteran street, Malang, Indonesia
* mas.ud[at]ub.ac.id
Abstract
Sugar cane (Saccharum officianarum) is an important commodity because it is widely used as raw material for sugar and MSG. Data on sugarcane production has not been used optimally, except for administrative purposes. The data, if used appropriately, can be used to predict the yield of sugar cane which can be utilized by cooperatives and farmers. This research was conducted to design an information system that can be used to forecast sugarcane yields in the working area of "KUD Subur Malang". The information system design process is carried out by implementing Machine Learning. The results of sugarcane yield forecasting using machine learning implementation in KUD Subur Malang showed the best results using the gradient boosting algorithm with 75% model accuracy. Website-based yield forecasting information system can be used as a production forecasting tool for KUD Subur to improve its business processes. Sugarcane forecasting information system can be well received by users.
Keywords: machine learning, sugarcane, gradient boosting
Topic: Information System and Technology