MonDT - Decision Trees for Classification with Monotonicity Constraints

This repository implements two different decision trees for monotonic classification:

• Rank Discrimination Measure Tree (RDMT). This decision tree can be built with three different rank discrimination measures: Gini, Shannon, or Pessimistic.

  Marsala, C., & Petturiti, D. (2015). Rank discrimination measures for enforcing monotonicity in decision tree induction. Information Sciences, 291, 143-171.

• Rank Entropy-Based Decision Trees for Monotonic Classification (REMT)

  Hu, Q., Che, X., Zhang, L., Zhang, D., Guo, M., & Yu, D. (2011). Rank entropy-based decision trees for monotonic classification. IEEE Transactions on Knowledge and Data Engineering, 24(11), 2052-2064.

Example

The file exampleMonDT is an example of the execution of these decision trees with the the Artiset data-set. The results obtained are the followings:

Algorithm	Accuracy	MAE	NMI
RDMT_Gini	0.81	0.19	0.0
RDMT_Shannon	0.86	0.14	0.0
RDMT_Pessimistic	0.88	0.12	0.0
REMT	0.86	0.14	0.0