A Novel Method of Parallel GPU Implementation of KNN Used in Text Classification

Automatic text classification is useful when websites have huge volume of web pages or other articles. K-Nearest Neighbour (KNN) is a way to classify the domains of text documents. The performance of text classification depends on lots of factors but KNN process contributes most of computational loads. We present a novel method of parallel GPU implementation of KNN with speed-ups of 40 times compared with CPU implementation.

[1]  Laxmikant V. Kalé,et al.  Highly scalable parallel sorting , 2010, 2010 IEEE International Symposium on Parallel & Distributed Processing (IPDPS).

[2]  Regina Berretta,et al.  GPU-FS-kNN: A Software Tool for Fast and Scalable kNN Computation Using GPUs , 2012, PloS one.

[3]  N. Altman An Introduction to Kernel and Nearest-Neighbor Nonparametric Regression , 1992 .

[4]  David J. Hand,et al.  Choosing k for two-class nearest neighbour classifiers with unbalanced classes , 2003, Pattern Recognit. Lett..

[5]  Michel Barlaud,et al.  Fast k nearest neighbor search using GPU , 2008, 2008 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops.

[6]  C. A. Murthy,et al.  On visualization and aggregation of nearest neighbor classifiers , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.