For each subspace of the predictors, we build a classification tree or logistic regression tree.
We also performed a new classification of 584 kinases on the basis of the similarity score which allowed us to build a "classification tree" of the currently known kinases ATP cavities.
To solve the class prediction problem, we built a classification tree on the learning set, and then sought the optimal placement of each test sample within the classification tree.
CART was used to build a classification tree with the appropriateness score of the panel ratings as a dependent variable, as categorical variable (appropriate, uncertain or inappropriate).
We also implemented this more general approach, as available in the "tree" package of the R statistics software (R Development Core Team 2006), using the standard recursive partitioning model (Foulkes et al. 2004) to build a classification tree.
Algorithm ?? describes the basic process of building a classification tree.
Factors included in a resulting tree, as well as its shape, are sensitive to the choice of criteria used to build and prune a classification tree, and this choice is often not straightforward.
To further explore how these features relate to H2A/H4R3me2s, we built a single classification tree [ 28], which, compared to the random forests ensemble of trees, may more readily reveal interpretable rules, albeit at the cost of decreased classification accuracy.
To trace the importance of variables responsible for the discrimination of both groups, a classification tree was built.
In parallel the same set of variables were analysed using a classification tree algorithm in order to build classification rules.
A classification tree will be grown for the chosen observations.
