Frequent Substring-Based Sequence Classification with an Ensemble of Support Vector Machines Trained Using Reduced Amino Acid Alphabets.
Charith ChitraranjanLoai Al NimerOmar Al AzzamSaeed SalemAnne M. DentonMuhammad J. IqbalShahryar F. KianianPublished in: ICMLA (2) (2011)
Keyphrases
- sequence classification
- amino acids
- sequence alignment
- support vector
- learning machines
- protein sequences
- string kernels
- svm classifier
- ensemble learning
- sequence data
- training set
- hidden markov models
- secondary structure
- computational biology
- markov models
- feature selection
- discriminative learning
- support vector machine
- protein structure
- conditional random fields
- maximum margin
- pairwise
- generalization ability
- loss function
- ensemble methods
- classification accuracy
- cross validation
- kernel function
- training data
- higher order
- data structure
- learning algorithm
- base classifiers
- training examples
- hyperplane
- binding sites
- feature space
- biological sequences