An Intelligent E -Nose System For Discrimination Of Alcholic Odorants

This paper presents a neural network classifier for classification of individual wine odors. The data set used for classification was obtained from already reported responses of thick -film tin oxide sensor array exposed to five different alcohol ic beverages. The proposed classifier was trained with back -propagation algorithm and fuzzy memberships were used as target vectors in the output feature space. Transformed Cluster Analysis (TCA) was used as a data pre -processing technique and it was obser ved that pre -processing the sensor output data with TCA significantly enhances the classification capability and the error performance of the proposed neuro -fuzzy classifier.