A preliminary experimental comparison of recursive neural networks and a tree kernel method for QSAR/QSPR regression tasks.
Alessio MicheliFilippo PorteraAlessandro SperdutiPublished in: ESANN (2004)
Keyphrases
- experimental comparison
- kernel methods
- recursive neural networks
- reproducing kernel hilbert space
- machine learning
- support vector
- drug discovery
- kernel function
- support vector machine
- feature selection
- regression model
- kernel matrix
- reproducing kernel
- tree kernels
- gesture recognition
- tree structure
- feature space
- rademacher complexity
- learning problems
- neural network
- similarity measure
- tree structures
- decision trees
- learning tasks
- active learning