Decision Tree Arithmetic Based on Conditional Probability
Lu Bai · Journal of Shenyang University · 2009
The basic principle of decision tree was introduced.In the process of building a decision tree when using ID3(Interactive Dichotomic Version 3) algorithm,according to the principle of maximum information gain to select attributes,the formula is more complex,the corresponding complexity is relatively high,and consuming hardware resources.Compared with the ID3 algorithm,the conditional probability decision tree algorithm selects the properties in accordance with the principle of maximum conditional probability.Conditional probability is able to link attributes and categories,the formula is relatively simple and effective to reduce the complexity of the calculation,and the establishment of decision tree improves the accuracy and detection speed of the system.