|
Electronic Nose For Black Tea Quality Evaluation Using Kernel Based Clustering ApproachKeywords: Kernel , Feature Space , Nonlinear Mapping , Electronic Nose , Black Tea , PCA , KPCA , LDA , KLDA. Abstract: Black Tea is conventionally tested by human sensory panel called “Tea Tasters”, who assignquality scores to different tea samples. This paper proposed a method of separation using thedevice named as electronic noise. The various tea samples have been analyzed using thepopular method of separation, like PCA and LDA. For better separation among different scores oftea samples, the kernel based PCA as well as kernel based LDA methods have been consideredin this case as the clustering algorithm. The method exhibits a better performance than those oftraditional methods. Also the separation index has been evaluated and shows its efficacy.
|