PERBANDINGAN PERFORMA ALGORITMA RANDOM FOREST REGRESSOR DAN XGBOOST REGRESSOR DALAM PREDIKSI BAND GAP MATERIAL SEMIKONDUKTOR SILIKON Muhammad Rizky Anugrah(a*) , Teguh Budi Prayitno (b*) , Haris Suhendar (c*)
a) Program Studi Fisika, Universitas Negeri Jakarta, Jakarta Timur 13220, Indonesia
*muhammadrizkyanugrah_179[at]gmail.com
b) Program Studi Fisika, Universitas Negeri Jakarta, Jakarta Timur 13220, Indonesia
*haris_suhendar[at]unj.ac.id,
c)Program Studi Fisika, Universitas Negeri Jakarta, Jakarta Timur 13220, Indonesia
*teguh-budi[at]unj.ac.id
Abstract
Semiconductor materials play a crucial role in various modern technological applications, including electronics, photovoltaics, and optoelectronics. One of the primary properties of semiconductor materials is the band gap, which is the energy required to excite an electron. The band gap is a key parameter influencing the electronic and optical behavior of semiconductor materials. In this study, machine learning is used to predict the band gap values of the semiconductor material silicon. Silicon is one of the most important and widely used semiconductor materials in various modern technological applications. The dataset used is taken from the Materials Project (MP), which provides a wide range of data on tested materials, including their features and characteristics. MP offers information on various types of materials, such as material properties, crystal structures, thermal stability, and others that can be used to build machine learning models. This research aims to develop and compare the performance of two machine learning algorithms, namely Random Forest Regressor and XGBoost Regressor, in predicting the band gap of silicon materials with high accuracy based on the features and characteristics available in the Materials Project dataset. This study also involves a comprehensive evaluation of both machine learning models in predicting the band gap of silicon materials.
