Horizontally grouping the training set is equivalent to Bagging integrated learning implemented after bootstrap sampling on t.
Bagging [6] and boosting [25] are two most popular examples, which work on bootstrap samples of the training set.
In order to analyze the hierarchical organization of the B-cell C3NET gene regulatory network, we perform a three-step procedure based on bootstrap samples of the data.
The individual trees are trained on bootstrap samples of the dataset, while the final predictions of a RF model are obtained by averaging over all the trees ('voting') to guard against overfitting the data.
Specifically, each tree is built on a bootstrap sample of the training dataset, so that the bootstrap sample contains, on average, 63% of the unique original samples [16], [18], [19].
The trees are randomised firstly by being based on separate bootstrap samples of the data pool, samples of N out of N objects chosen with replacement.
At the same time, to keep correlation low when trees are grown we use randomization, such as each tree is grown on a bootstrap sample of the training set and the number of predictors in each specified tree is much smaller than the total number of total variables in the training set.
Each tree is formed on a bootstrap sample of the individuals (training dataset), while individuals that are not in the bootstrap sample (out-of-bag samples = OOB), are used for estimation of the mean squared error of prediction.
The model was based on 1000 bootstrap samples of the training set, with 1000 classification trees generated with a view to classifying cases as "inflammatory" and "not-inflammatory" on the basis of microarray gene expression measurements.
Specifically, each single tree is fully grown on a bootstrap sample of the training data, without pruning the leaves, and at each node split only a subset of the variables is considered.
