Regularized linear classifiers have been successfully applied in undersampled, i.e. small sample size/high dimensionality biomedical classification problems. Additionally, a desig...
Spectral classification, segmentation and data reduction are the three main problems in hyperspectral image analysis. In this paper we propose a Bayesian estimation approach which ...
Nadia Bali, Ali Mohammad-Djafari, Adel Mohammadpou...
FaCT (Fast Classification of Terminologies) is a Description Logic (DL) classifier that can also be used for modal logic satisfiability testing. The FaCT system includes two reason...
: In this paper we present our work on applying Belief Augmented Frames to the text classification problem. We formulate the problem in two alternative ways, and we evaluate the pe...
This paper shows how linguistic classification knowledge can be extracted from numerical data for pattern classification problems with many continuous attributes by genetic algori...