2013 | OriginalPaper | Buchkapitel
CLUEKR : CLUstering Based Efficient kNN Regression
verfasst von : Harshit Dubey, Vikram Pudi
Erschienen in: Advances in Knowledge Discovery and Data Mining
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
K
-Nearest Neighbor based regression algorithm assigns a value to the query instance based on the values of its neighborhood instances. Although
k
NN has proved to be a ubiquitous classification/regre-ssion tool with good scalability but it suffers from some drawbacks. One of its biggest drawback is that, it is a lazy learner i.e. it uses all the training data at runtime. In this paper, we propose a novel, efficient and accurate, clustering based
k
NN regression algorithm CLUEKR having the advantage of low computational complexity. Instead of searching for nearest neighbors directly in the entire dataset, we first hierarchically cluster the data and then find the cluster in which the query point should lie. Our empirical experiments with several real world datasets show that our algorithm reduces the search space for
k
NN significantly and is yet accurate.