Making large-scale support vector machine learning practical
Advances in kernel methods
Locally Adaptive Metric Nearest-Neighbor Classification
IEEE Transactions on Pattern Analysis and Machine Intelligence
Automatic Textual Document Categorization Based on Generalized Instance Sets and a Metamodel
IEEE Transactions on Pattern Analysis and Machine Intelligence
RCV1: A New Benchmark Collection for Text Categorization Research
The Journal of Machine Learning Research
Pattern Recognition Using Average Patterns of Categorical k-Nearest Neighbors
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 4 - Volume 04
SVM-KNN: Discriminative Nearest Neighbor Classification for Visual Category Recognition
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 2
Decompositional Rule Extraction from Support Vector Machines by Active Learning
IEEE Transactions on Knowledge and Data Engineering
An adaptable k-nearest neighbors algorithm for MMSE image interpolation
IEEE Transactions on Image Processing
Neighbor-weighted K-nearest neighbor for unbalanced text corpus
Expert Systems with Applications: An International Journal
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
A comparison of methods for multiclass support vector machines
IEEE Transactions on Neural Networks
Face Recognition Using Total Margin-Based Adaptive Fuzzy Support Vector Machines
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
The support vector machine (SVM) has proved effective in classification. However, SVM easily becomes intractable in its memory and time requirements to deal with the large data, and also can not nicely deal with noisy, sparse, and imbalanced data. To overcome these issues, this paper presents a new local support vector machine that first finds k nearest neighbors from each class respectively for the query sample and then SVM is trained locally on all these selected nearest neighbors to perform the classification. This approach is efficient, simple and easy to implement. The conducted experiments on challenging benchmark data sets validate the proposed approach in terms of classification accuracy and robustness.