Assessing the ability to predict human percepts of odor quality from the detector responses of a conducting polymer composite-based electronic nose

Abstract

The responses of a conducting polymer composite “electronic nose” detector array were used to predict human perceptual descriptors of odor quality for a selected test set of analytes. The single-component odorants investigated in this work included molecules that are chemically quite distinct from each other, as well as molecules that are chemically similar to each other but which are perceived as having distinct odor qualities by humans. Each analyte produced a different, characteristic response pattern on the electronic nose array, with the signal strength on each detector reflecting the relative binding of the odorant into the various conducting polymer composites of the detector array. A “human perceptual space” was defined by reference to English language descriptors that are frequently used to describe odors. Data analysis techniques, including standard regression, nearest-neighbor prediction, principal components regression, partial least squares regression, and feature subset selection, were then used to determine mappings from electronic nose measurements to this human perceptual space. The effectiveness of the derived mappings was evaluated by comparison with average human perceptual data published by Dravnieks. For specific descriptors, some models provided cross-validated predictions that correlated well with the human data (above the 0.60 level), but none of the models could accurately predict the human values for more than a few descriptors.