Linguistic fuzzy classification of olfactive signals

This paper presents a method for fuzzy classification of odour samples that are obtained from an electronic nose which uses conducting polymer sensors. These sensors are very sensitive but not specific. Linguistic expressions describing the responses of both individual sensors and the sensor array to each chemical are derived from a fuzzy model of the sensor data. A small scale application of the method in which eight sensors were used to classify three odours yielded 100% classification accuracy on out-of-sample data