The random forest algorithm under the MapReduce framework has too many redundant and
irrelevant features, low training feature information, and low parallelization efficiency when …