QCM based Electronic Nose for Black Tea Quality Evaluation by Different Data Analysis Techniques
Moumita Guha1, BipanTudu2, Pritam Singha Roy3
1Moumita Guha, Department of Instrumentation, Jadavpur University, Kolkata (West Bengal), India.
2BipanTudu, Department of Instrumentation, Jadavpur University, Kolkata (West Bengal), India.
3Pritam Singha Roy, Research Scholar, Pacific University, (Rajasthan), India.
Manuscript received on 06 June 2019 | Revised Manuscript received on 30 June 2019 | Manuscript Published on 04 July 2019 | PP: 794-797 | Volume-8 Issue-1S4 June 2019 | Retrieval Number: A11460681S419/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The common and effective technique to find the quality of tea is by sensor activities of the human called Tea Tasters. But this technique is most predictable in nature. For this reason very sophisticated sensor network has been designed. Electronic Nose is the most efficiently used to calculate the behavior and quality of black tea. An Electronic Nose sensor has been developed, consisting of an array of five AT-CUT 10 MHz QCM sensors. The array has been exhibits by the aroma characteristics of different types of black tea (CTC, Orthodox) and the response has been monitored online though Data Acquisition System. The data obtained in this manner has been clustered with the help of different algorithm(PCA,LDA). A BPMLP network has also been incorporated for the purpose of classification. The paper is concluded with comparison of cluster validity indices of original data with those of clusters obtained with different algorithms and the performance of the system which is very effective, is justified.
Keywords: Black Tea; Electronic Nose; QCM; Clustering; Classification; PCI; LDA; ICA; BPMLP; Dunn’s Index.
Scope of the Article: Data Analytics