KNN Euclidean Distance Model Performance on Aquilaria Malaccensis Oil Qualities

Abstract

Agarwood is a highly prized and useful forest product. In Southeast Asia, Aquilaria Malaccensis species are typically the most prevalent. This agarwood is usually used in the manufacture of medicine, the production of high-quality perfumes, and is also used in religious and ethnic ceremonies. According to the study, the agarwood grading process entirely relies on human senses. The graders will evaluate the agarwood oil's color concentration with their unaided eyes and evaluate the amount of scent emitted with their noses. This approach has been proven to have several limitations, including that the graders' health will suffer, the grading procedure will take a very long time, and will consume high operating expenses. Therefore, an established standard grading model that is faster, easier, and more accurate needs to be introduced. Previous researchers found that chemical compounds contained in agarwood oil can be used to grade the quality of agarwood oil. Therefore, this study has used the data obtained that contains significant chemical compounds as input to develop a grading model with the support of machine learning and artificial intelligence, which is the k-Nearest Neighbor (KNN) technique. The output of this grading model is the classification of agarwood oil according to its quality, which includes four different grades. The results of the implementation of KNN grading of this model found that this model has very excellent performance by obtaining 100% for each measurement for the performance evaluator of the classifier.