Bi-Parameter Space Partition for Cost-Sensitive SVM

Authors: Bin Gu, Victor S. Sheng, Shuo Li

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results on seven normal datsets and four imbalanced datasets, show that our proposed method has better generalization ability and than various kinds of grid search methods, however, with less running time.
Researcher Affiliation Collaboration Bin Gu , Victor S. Sheng , Shuo Li , School of Computer & Software, Nanjing University of Information Science & Technology, China Department of Computer Science, University of Western Ontario, Canada Department of Computer Science, University of Central Arkansas, Arkansas GE Health Care, Canada jsgubin@nuist.edu.cn, ssheng@uca.edu, Shuo.Li@ge.com
Pseudocode Yes Algorithm 1 CR-BPSP (CRs-based BPSP algorithm)
Open Source Code No The paper states 'We implemented SPη+GSλ, SPλ+GSη, and our BPSP in MATLAB.' but does not provide any link or explicit statement about the code being open-source or publicly available.
Open Datasets Yes The sonar (Son), ionosphere (Ion), diabetes (Dia), breast cancer (BC), heart (Hea), and hill-valley (HV) datasets were obtained from the UCI benchmark repository [Bache and Lichman, 2013]... Ecoli1, Ecoli3, Vowel0, and Vehicle0 are the imbalanced datasets from the KEEL-dataset repository2. http://sci2s.ugr.es/keel/imbalanced.php.
Dataset Splits Yes We selected 30% from a dataset once as a validation set. The validation set was used with a 5-fold CV procedure to determine the optimal parameters.
Hardware Specification Yes All experiments were performed on a 2.5-GHz Intel Core i5 machine with 8GB RAM and MATLAB 7.10 platform.
Software Dependencies Yes All experiments were performed on a 2.5-GHz Intel Core i5 machine with 8GB RAM and MATLAB 7.10 platform.
Experiment Setup Yes Design of Experiments We compare the generalization ability and runtime of BPSP with other three typical model selection methods of CS-SVM: (1) grid search (GS): a two-step grid search strategy is used for 2C-SVM. The initial search is done on a 20 20 coarse grid linearly spaced in the region {(log2 C+, log2 C )| 9 log2 C+ 10, 9 log2 C 10}, followed by a fine search on a 20 20 uniform grid linearly spaced by 0.1... C( , +) was set to 2, 5, 10... Gaussian kernel K(x1, x2) = exp( κ x1 x2 2) was used in all the experiments with κ {10 3, 10 2, 10 1, 1, 10, 102, 103}...