MSVM modelling on agarwood oil various qualities classification

The aroma of agarwood oil is frequently connected with wealth and prestige. Because of its aroma, the oil is in great demand all over the world. Unfortunately, no standard method for classifying the quality of agarwood oil has been proposed to the world. This is essential to preserving the quality of agarwood oil and preventing fraudulent quality in the market. With that, for agarwood oil grading, a quality classification model must be developed. As part of the ongoing study for this standard’s model development, an intelligent algorithm function has been implemented to ensure the model’s capability is completely unquestionable. The intelligent algorithm used was Support Vector Machine (SVM) as main structure model. Then, the model has been added with Multiclass Classifier function by using One Verses One (OVO) strategy. The previous researcher's data was used in the analysis process, which included four classes of agarwood oil quality samples: low, medium low, medium high, and high quality. The result was a quality categorization of low, medium low, medium high or high quality, whereas the input was chemical abundances (percentages). The combination of both functions produced extremely good results for agarwood oil quality classification. The simulation platform was MATLAB software version r2020a, which was used for the desk study. The findings of this study will undoubtedly be useful in future agarwood oil research, particularly in the grade categorization section.