Cong Gao
University of Regina
12 Papers
53 Citations
Cong Gao is an academic researcher from University of Regina. The author has contributed to research in topics: Support vector machine & Boundary (topology). The author has an hindex of 8, co-authored 12 publications.
Chat about Author
Papers
A weighted one-class support vector machine
TL;DR: The experimental results demonstrate that WOC-SVM outperforms the standard OC-S VM when using the proposed instance-weighted strategy, and performs better than previous ones.
73
Granularity-driven sequential three-way decisions: A cost-sensitive approach to classification
TL;DR: Efforts to develop a granularity-driven sequential three-way decision model to address the construction of granules or costs in the decision process, and handle dual-constraint satisfaction problems considering both decision process and decision result costs are described.
70
Actionable strategies in three-way decisions
TL;DR: Four change-based actionable models according to action benefit and action cost are presented and the bounds of the cost and benefit are provided and the maximum benefit under limited cost and the minimum cost for a desired benefit are quantified.
49
Neighbors' distribution property and sample reduction for support vector machines
Fa Zhu,Fa Zhu,Jian Yang,Ning Ye,Cong Gao,Guobao Li,Tongmin Yin +6 more
- 01 Mar 2014
TL;DR: A new method is proposed to find the boundary regions of each class instead of overlap regions, which can solve the problem, which the methods based on finding overlap regions cannot deal with.
27
Finding the samples near the decision plane for support vector learning
TL;DR: The extended nearest neighbor chain is proposed to find samples near the decision plane to avoid degrading performance in the cases of no overlapping region between different classes.
26