Inductive logic programming: Difference between revisions

Revision as of 00:27, 8 June 2007

Revision as of 04:51, 21 June 2007

Line 1:

'''Inductive logic programming''' ('''ILP''') is a subfield of [[machine learning]] approach which uses techniques of [[logic programming]]. Fromas a databaseuniform ofrepresentation factsfor examples, background knowledge and expectedhypotheses. results,Given whichan areencoding dividedof intothe positiveknown background knowledge and negativea set of examples represented as a logical database of facts, an ILP system triesderives toan derive ahypothesised logic program thatwhich provesentails all the positive and none of the negative examples.

Schema: ''positive examples'' + ''negative examples'' + ''background knowledge'' => ''ruleshypothesis''.

Inductive logic programming is particularly useful in [[bioinformatics]] and [[natural language processing]].