The Grading of Agarwood Oil Quality Based on Multiclass Support Vector Machine (MSVM) Model

• Published in: 2022 IEEE 18th International Colloquium on Signal Processing and Applications, CSPA 2022 - Proceeding
• Main Author: Mohd Amidon A.F.; Zawani Mahabob N.; Hazwa Mohd Huzir S.M.; Mohd Yusoff Z.; Ismail N.; Taib M.N.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85132759329&doi=10.1109%2fCSPA55076.2022.9782021&partnerID=40&md5=8e892b5de945efd5864a3bb650a18acf

Agarwood oil is one of the most valued oils among the world's peoples, which contributes to its ever-increasing demand. It has a variety of advantages and applications, including the manufacturing of incense and fragrances, and also is employed in traditional medicine. However, without a standard grading model for agarwood oil has resulted in certain flaws in the grading procedure. To address these flaws, a standard grading model must be developed and deployed as soon as possible. By continuing the research study of standard grading model development, intelligent algorithm must be implemented as main function to establishment of this standard to ensure that the model's capability is entirely unquestioned. One of classification algorithm which is Support Vector Machine algorithm has been chosen and multiclass classifier algorithm has been used as supporter to SVM. One of multiclass classifier strategies which is One versus All strategy has been implemented to improve the ability of SVM. By combining both intelligent techniques, the model was able to be function as multiclass classification model, known as Multiclass Support Vector Machine (MSVM) model. In MSVM model, percentage of abundance chemical compounds have been used as input and quality (low, medium low, medium high and high) was used as output. The Matlab software version r2020a was used in this research work to train and test the model's performance. The results revealed that the model passed the performance requirements standard while employing the multiclass function. The findings of this study will undoubtedly be useful in future agarwood oil research, particularly in quality categorization. © 2022 IEEE.