Cost-Sensitive Decision Trees with Completion Time Requirements

In many classification tasks, managing costs and completion times are the main concerns. In this paper, we assume that the completion time for classifying an instance is determined by its class label, and that a late penalty cost is incurred if the deadline is not met. This time requirement enriches the classification problem but posts a challenge to developing a solution algorithm. We propose an innovative approach for the decision tree induction, which produces multiple candidate trees by allowing more than one splitting attribute at each node. The user can specify the maximum number of candidate trees to control the computational efforts required to produce the final solution. In the tree-induction process, an allocation scheme is used to dynamically distribute the given number of candidate trees to splitting attributes according to their estimated contributions to cost reduction. The algorithm finds the final tree by backtracking. An extensive experiment shows that the algorithm outperforms the top-down heuristic and can effectively obtain the optimal or near-optimal decision trees without an excessive computation time.