Analysis of the Effects of Unexpected Outliers in the Classification of Spectroscopy Data
Glavin, Frank G.
Madden, Michael G.
MetadataShow full item record
This item's downloads: 826 (view details)
Glavin, F. G., & Madden, M. G. (2009). Analysis of the Effects of Unexpected Outliers in the Classifcation of Spectroscopy Data. Paper presented at the 20th Irish Conference on Artificial Intelligence and Cognitive Science
Multi-class classification algorithms are very widely used, but we argue that they are not always ideal from a theoretical perspective, because they assume all classes are characterised by the data, whereas in many applications, training data for some classes may be entirely absent, rare, or statistically unrepresentative. We evaluate one- sided classifiers as an alternative, since they assume that only one class (the target) is well characterised. We consider a task of identifying whether a substance contains a chlorinated solvent, based on its chemical spectrum. For this application, it is not really feasible to collect a statistically representative set of outliers, since that group may contain anything apart from the target chlorinated solvents. Using a new one-sided classification toolkit, we compare a One-Sided k-NN algorithm with two well- known binary classification algorithms, and conclude that the one-sided classier is more robust to unexpected outliers.