Lazy Learners at work: the Lazy Learning Toolbox
Gianluca Bontempi, Mauro Birattari, Hugues Bersini, Hans-Jürgen Zimmerman · 1999
Lazy Learning is a memory-based technique that, once a query is received, extracts a prediction interpolating locally the neighboring examples of the query which are considered relevant according to a distance measure. In previous works, we presented a Lazy Learning method which selects automatically on a query-by-query basis the optimal number of neighbors to be considered for each prediction. This learning method proved to be a very eective technique in a number of academic and industrial case studies, ranging from time series prediction to data modeling and nonlinear control. This paper discusses the implementation of the Lazy Learning technique in a toolbox for use with Matlab c and its successful application to the problem of multivariate regression proposed by the Third Erudit Competition. 1 Introduction Lazy learning (Aha, 1997) is a local learning technique which postpones all the computation until an explicit request for a prediction is received. The request is fullled ...