Knowledge based decision tree construction with feature importance domain knowledge
Md. Ridwan Al Iqbal, Saiedur Rahman, Syed Irfan Nabil, Ijaz Ul Amin Chowdhury · 2012
Decision Tree is a widely used supervised learning algorithm due its many advantages like fast non parametric learning, comprehensibility and son. But, Decision Tree require large training set to learn accurately because, decision tree algorithms recursively partition the data set that leaves very few instances in the lower levels of the tree. In order to address this drawback, we present a novel algorithm named Importance Aided Decision Tree (IADT). It that takes Feature Importance as an additional domain knowledge. Additional domain knowledge have been shown to enhance the performance of learners. Decision Tree algorithm always finds the most important attributes in each node. Thus, Importance of features is a relevant domain knowledge for decision tree algorithm. Our algorithm uses a novel approach to incorporate this feature importance score into decision tree learning. This approach makes decision trees more accurate and robust. We presented theoretical and empirical performance analysis to show that IADT is superior to standard decision tree learning algorithms.