Using Prototype-Based Classification for Automatic Knowledge Acquisition

In this paper we describe how prototype-based classification can be used for knowledge acquisition in image classification. We describe the necessary functions a prototype-based classifier should have. These functions need to be developed for a low number of samples and for a large number of samples; they are feature subset selection, similarity learning and prototype selection. The classifier is applied to the internal mitochondrial movement of cells as an example. The aim was to discover the different dynamic signatures of mitochondrial movement. Our results and an outlook on future work are presented here.