Iterative Split Adjustment for Building Multilabel Decision Trees

A decision tree induction method for multilabel classification tasks (IS-MLT) is presented which uses an iterative approach for determining the best split at each node. The proposed method is applied to the task of identifying gene functional classifications of microarray data and the results for three datasets are compared to an existing multi-label tree approach [2, 22], and on one dataset to other published results. The results are compared using the micro-label F-measure, and overall IS-MLT is found to perform better.

• Type: Journal
• Year: 2010
• Where: BICOB
• Authors: Aiyesha Ma, Ishwar K. Sethi
• Keywords: BICOB 2010, Computational Biology, Gene Functional Classifications, Multilabel Classification Tasks, Tree Induction Method