Predictive Activity Profiling of Drugs by Topological-Fragment-Spectra-Based Support Vector Machines.
Kentaro KawaiSatoshi FujishimaYoshimasa TakahashiPublished in: J. Chem. Inf. Model. (2008)
Keyphrases
- support vector
- learning machines
- large margin classifiers
- logistic regression
- generalization ability
- classification accuracy
- multi class
- principal component analysis
- kernel function
- hyperplane
- human activities
- kernel methods
- cross validation
- topological properties
- radial basis function
- drug design
- loss function
- support vector machine
- topological spaces
- connected components
- decision function
- generalization bounds
- activity theory
- maximum margin
- predictive model
- support vector regression
- learning algorithm
- feature selection
- social networks