ILP for Bootstrapped Learning: A Layered Approach to Automating the ILP Setup Problem
Published in Nineteenth International Conference on Inductive Logic Programming (ILP'09), Leuven, Belgium, 2009
Recommended citation: S. Natarajan, G. Kunapuli, C. O' Reilly, R. Maclin, T. Walker, D. Page and J. W. Shavlik. ILP for Bootstrapped Learning: A Layered Approach to Automating the ILP Setup Problem. Nineteenth International Conference on Inductive Logic Programming (ILP'09), Leuven, Belgium, July 2-4, 2009. http://gkunapuli.github.io/files/09ilp.pdf
This paper introduces a new type of application for ILP called Bootstrapped Learning (BL). BL brings several challenges to ILP, including the need to automate the “ILP Set-Up” problem; small numbers of examples, in some cases no negative examples; and the need to bootstrap or to automatically base learning in part on the results of earlier learning sessions. The paper introduces BL, discusses the general challenges it raises for ILP, and presents initial results in ILP for BL.