NP chunking using ILP

This is to report the results of approaching the problem of NP chunking using Inductive Logic Programming techniques. The problem, as de-ned in (Ramshaw and Marcus, 1995), is the machine learning of rules that recognise non-recursive, base NPs in text annotated with part-of-speech tags, by tagging each word as beingìnside' oròutside' an NP. (Consecutive NPs are appropriately treated.) The same input data as in the original experiment is used here, but the machine learning technique is Inductive Logic Programming, and more speciically the Progol algorithm. The problem is formulated as the machine learning of a Prolog predicate that will accept a part-of-speech tagged word and its context as input and associate it with the appropriate syntactic tag.