Learning from Sparse Data by Exploiting Monotonicity Constraints
Eric Altendorf, Angelo Restificar, Thomas Dietterich
When training data is sparse, more domain knowledge must be incorporated into the learning algorithm in order to reduce the effective size of the hypothesis space. This paper builds on previous work in which knowledge about qualitative monotonicities was formally represented and incorporated into learning algorithms (e.g., Clark & Matwin's work with the CN2 rule learning algorithm). We show how to interpret knowledge of qualitative influences, and in particular of monotonicities, as constraints on probability distributions, and to incorporate this knowledge into Bayesian network learning algorithms. We show that this yields improved accuracy, particularly with very small training sets (e.g. less than 10 examples).
PDF Link: /papers/05/p18-altendorf.pdf
AUTHOR = "Eric Altendorf
and Angelo Restificar and Thomas Dietterich",
TITLE = "Learning from Sparse Data by Exploiting Monotonicity Constraints",
BOOKTITLE = "Proceedings of the Twenty-First Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-05)",
PUBLISHER = "AUAI Press",
ADDRESS = "Arlington, Virginia",
YEAR = "2005",
PAGES = "18--26"