Data Mining Using Support Vector Machines

The objectives of this work are; first to propose BbSVM and BmSVM as new boosting algorithms for enhancing the accuracy and performance of common SVM. Second,to show the robustness of various kind of kernels for BbSVM and BmSVM classifiers,and a comparison of different constructing methods for Multi...

Full description

Saved in:
Bibliographic Details
Main Author: Chamasemani, Fereshteh Falah
Format: Thesis
Published: 2011
Tags: Add Tag
No Tags, Be the first to tag this record!