A balanced iterative random forest algorithm is proposed to select the most relevant genes to the disease and can be used in the classification and prediction process.
Balanced iterative random forest is an embedded feature selector that follows a backward elimination approach. The base learning algorithm is random forest which is involved in the process of determining which features are removed at each step. The algorithm starts with the entire set of features in the dataset. At every iteration, the number of the attributes is reduced by removing those attributes which have zero importance value. After discarding those genes, a new random forest is built again with the selected set of genes that yields the smallest out-of-bag (OOB) error rate
BIRF
A Balanced Iterative Random Forest For Gene Selection From Microarray
Brought to you by:
aanaissi
Downloads:
0 This Week