An artificial neural network is capable of predicting odour intensity

For the first time, an artificial neural network (ANN) has been employed for predicting the intensity of gas mixtures comprising different odour components. Sensory assessments are necessary but they are time-consuming, harmful, and expensive. Therefore, an instrumental quantification of subjective sensory assessments is highly desired. Because of nonlinearities arising in sensory-instrumental relationships, we decided for an ANN that was trained by gas chromatographic signals of gas mixtures. The ANN could be demonstrated to classify odour intensity fairly well.