Effect of training data size and noise level on support vector machines virtual screening of genotoxic compounds from large compound libraries.
Pankaj KumarXiao Hua MaXianghui LiuJia JiaBu-Cong HanYing XueZe-Rong LiSheng-Yong YangYu-Quan WeiYu Zong ChenPublished in: J. Comput. Aided Mol. Des. (2011)
Keyphrases
- noise level
- drug discovery
- training data
- virtual screening
- support vector
- noisy images
- noise reduction
- noise estimation
- error rate
- support vector machine
- chemical structures
- chemical compounds
- decision trees
- data sets
- learning algorithm
- supervised learning
- training set
- early stage
- image processing
- feature selection
- similarity searching
- edge detection
- kernel function
- data analysis
- multimedia
- data mining