Name
Abstract | ||
|---|---|---|
KNN is one of the most popular data mining methods for classification, but it often fails to work well with inappropriate choice of distance metric or due to the presence of numerous class-irrelevant features. Linear feature transformation methods have been widely applied to extract class-relevant information to improve kNN classification, which is very limited in many applications. Kernels have also been used to learn powerful non-linear feature transformations, but these methods fail to scale to large datasets. In this paper, we present a scalable non-linear feature mapping method based on a deep neural network pretrained with Restricted Boltzmann Machines for improving kNN classification in a large-margin framework, which we call DNet-kNN. DNet-kNN can be used for both classification and for supervised dimensionality reduction. The experimental results on two benchmark handwritten digit datasets and one newsgroup text dataset show that DNet-kNN has much better performance than large-margin kNN using a linear mapping and kNN based on a deep autoencoder pretrained with Restricted Boltzmann Machines. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/ICDM.2009.27 | ICDM |
Keywords | Field | DocType |
non-linear dimensionality reduction,large-margin knn classification,deep neural network,boltzmann machines,non-linear feature mapping,handwritten digit datasets,restricted boltzmann machines,dnet-knn,linear feature transformation method,pattern classification,knn classification,scalable non-linear feature mapping,linear feature transformation methods,deep learning,powerful non-linear feature transformation,class-relevant information extraction,deep non-linear feature mapping,deep autoencoder,large-margin knn,benchmark handwritten digit datasets,rbm,deep neural networks,large margin,data mining,newsgroup text dataset,numerous class-irrelevant feature,neural nets,boltzmann machine,data models,neural network,feature extraction,stochastic processes,artificial neural networks | Data mining,Data modeling,Boltzmann machine,Dimensionality reduction,Computer science,Metric (mathematics),Artificial intelligence,Deep learning,Artificial neural network,Autoencoder,Pattern recognition,Feature extraction,Machine learning | Conference |
ISSN | ISBN | Citations |
1550-4786 E-ISBN : 978-0-7695-3895-2 | 978-0-7695-3895-2 | 28 |
PageRank | References | Authors |
1.23 | 18 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Renqiang Min | 1 | 43 | 3.53 |
| David A. Stanley | 2 | 28 | 1.56 |
| Zineng Yuan | 3 | 48 | 2.44 |
| Anthony J. Bonner | 4 | 733 | 422.63 |
| Zhaolei Zhang | 5 | 237 | 19.25 |