FRPS: A Fuzzy Rough Prototype Selection method

•
• Prototype Selection selects high-quality instances to improve k NN classification.
•
• State-of-the-art prototype are accurate but generally slow.
•
• We propose a Prototype Selection method based on fuzzy rough set theory.
•
• Experimental results show that this method is fast and significantly more accurate.

The k Nearest Neighbour (k NN) method is a widely used classification method that has proven to be very effective. The accuracy of k NN can be improved by means of Prototype Selection (PS), that is, we provide k NN with a reduced but reinforced dataset to pick its neighbours from. We use fuzzy rough set theory to express the quality of the instances, and use a wrapper approach to determine which instances to prune. We call this method Fuzzy Rough Prototype Selection (FRPS) and evaluate its effectiveness on a variety of datasets. A comparison of FRPS with state-of-the-art PS methods confirms that our method performs very well with respect to accuracy.