Enhancing Aquilaria species classification using agarwood oil analysis and k-Nearest Neighbors (KNN) as machine learning / Amir Hussairi Zaidi ... [et al.]

Aquilaria species is renowned for the aromatic resinous wood also known as agarwood and 17 accepted species are known can produce it. This tree has been long prized in various cultural and commercial context for its aromatic resinous wood, esteemed for its rich fragrance and versatile applications. The accurate classification of this agarwood-producing plant from Aquilaria species is essential for purposes such as conservation, sustainable resource management, trade regulation, research, and the preservation of cultural and economic traditions. This study employs machine learning, specifically the k-Nearest Neighbors (kNN) algorithm, to classify Aquilaria species based on chemical compounds features extracted from agarwood oil. Agarwood from four Aquilaria (A.) species: A. beccariana, A. crassna, A. malaccensis, and A. subintegra, is being used and the chemical compound is being analysed using the Gas Chromatography-Flame Ionization Detector (GC-FID). Subsequently, classifier performance is assessed using a confusion matrix to measure accuracy. The study not only demonstrates the effectiveness of this technique but also highlights its potential for future research related to Aquilaria and agarwood, reinforcing its relevance and applicability.