Indeed, electronic noses are instruments that consist of semiconductor gas sensors that respond by changing resistance. The incrementally-different sensors form a sensory array that first responds to a wide variety of chemical classes (odor compounds) when a sample passes over the array (Wilson & Baietto, 2009). The sensory array then sends data to an artificial neural network (ANN) to classify the sample (Mumyakmaz & Karabacak, 2013). The ANN relies on “training” to calibrate the electronic nose by building up pattern recognition based on this data; training consists of sampling several different known classes multiple times such that the ANN records and stores the change in resistances (Wilson & Baietto, 2009). As a result, the ANN uses the data in identifying
Indeed, electronic noses are instruments that consist of semiconductor gas sensors that respond by changing resistance. The incrementally-different sensors form a sensory array that first responds to a wide variety of chemical classes (odor compounds) when a sample passes over the array (Wilson & Baietto, 2009). The sensory array then sends data to an artificial neural network (ANN) to classify the sample (Mumyakmaz & Karabacak, 2013). The ANN relies on “training” to calibrate the electronic nose by building up pattern recognition based on this data; training consists of sampling several different known classes multiple times such that the ANN records and stores the change in resistances (Wilson & Baietto, 2009). As a result, the ANN uses the data in identifying