Classification of Agarwood Oil Using an Electronic Nose

Abstract

Presently, the quality assurance of agarwood oil is performed by sensory panels which has significant drawbacks in terms of objectivity and repeatability. In this paper, it is shown how an electronic nose (e-nose) may be successfully utilised for the classification of agarwood oil. Hierarchical Cluster Analysis (HCA) and Principal Component Analysis (PCA), were used to classify different types of oil. The HCA produced a dendrogram showing the separation of e-nose data into three different groups of oils. The PCA scatter plot revealed a distinct separation between the three groups. An Artificial Neural Network (ANN) was used for a better prediction of unknown samples.

Mathematical modeling for extraction of essential oil from Aquilaria crassna by hydrodistillation and quality of agarwood oil

“…Hidayat et al (2010) reported that agarwood oil grading has been traditionally performed by trained human graders (sensory panels). The disadvantages of this technique include subjectivity, poor reproducibility, time consumption and large labor expense (Keller, 1999).…”

“…The disadvantages of this technique include subjectivity, poor reproducibility, time consumption and large labor expense (Keller, 1999). A human nose cannot tolerate a high number of samples and work in continuous production because it fatigues rapidly with increasing number of samples (Hidayat et al, 2010;Keller, 1999). Identification of agarwood oil would seem to be impossible without the use of laboratory techniques (Barden et al, 2000).…”

Pollutant Recognition Based on Supervised Machine Learning for Indoor Air Quality Monitoring Systems

“…This transformation is defined in such a way that the first principal component has as high a variance as possible and each succeeding component in turn has the highest variance possible under the constraint that it be orthogonal to the preceding components. It is therefore known as an unsupervised data analysis method or algorithm since it "ignores" class labels [50][51][52]. In this research, PCA was used to remove any redundancy between the components of the projected vectors and reduce the dimension of the original dataset.…”

Identification of Fake Green Tea by Sensory Assessment and Electronic Tongue

“…In addition, the trained tea tasters cannot tolerate large numbers of samples because they fatigue rapidly with increasing number of samples (Hidayat et al, 2010).…”