The n-tuple Classifier: Too Good to Ignore

The n-tuple pattern recognition method has been tested using a selection of 11 large data sets from the European Community StatLog project, so that the results could be compared with those reported for the 23 other algorithms the project tested. The results indicate that this ultra-fast memory-based method is a viable competitor with the others, which include optimisation-based neural network algorithms, even though the theory of memory-based neural computing is less highly developed in terms of statistical theory. 1 Introduction A popular style of neural computation is to apply optimisation techniques to suitably designed neural network models. This has the advantages of good performance and reasonably firm theoretical underpinnings, but often suffers from hefty computational requirements. An alternative style is based on memorisation of randomly selected features. Although the theory is less well-developed, such methods offer an overwhelming advantage in computation speed. One of the..