This research aims to study the decision tree algorithm for data classification, data by random mushrooms out of 100, 500 and 1000, and then importing the Random Forest algorithm is random number of trees, and then view the value of the attribute that caused the incorporation of the node and leaf in the forest, random. If any attributes with values ranging from 0-NN0D99 to remove the attribute from the data, and then import data into decision tree algorithm to compare the size of the tree and the accuracy. By the results of the experiment showed that when random information that tree 100 data can reduce the size of, and the value must not be reduced. If the decision tree smaller will reduce the complexity of the data. The result is high accuracy value.
การแปล กรุณารอสักครู่..
