Inductive Logic Programming (ILP), Link Discovery (LD), Evidence Extraction and Link Discovery (EELD)

Inductive logic programming (ILP) is a subfield of machine learning which uses logic programming as a uniform representation for examples, background knowledge and hypotheses. Given an encoding of the known background knowledge and a set of examples represented as a logical database of facts, an ILP system will derive a hypothesised logic program which entails all the positive and none of the negative examples.

Schema: positive examples + negative examples + background knowledge => hypothesis.

Inductive logic programming is particularly useful in bioinformatics and natural language processingEhud Shapiro laid the theoretical foundation for inductive logic programming[1][2] and built its first implementation (Model Inference System) in 1981:[3] a Prolog program that inductively inferred logic programs from positive and negative examples. The term Inductive Logic Programming was first introduced[4] in a paper by Stephen Muggleton in 1991.[5] The term “inductive” here refers to philosophical (i.e. suggesting a theory to explain observed facts) rather than mathematical (i.e. proving a property for all members of a well-ordered set) induction.

Link discovery (LD) is an important task in data mining for counter-terrorism and is the focus of DARPA’s Evidence Extraction and Link Discovery
(EELD) research program. Link discovery concerns the identification of complex
relational patterns that indicate potentially threatening activities in large amounts of relational data. Most data-mining methods assume data is in the form of a feature-vector (a single relational table) and cannot handle multi-relational data.
Inductive logic programming is a form of relational data mining that discovers rules in first-order logic from multi-relational data. This paper discusses the application of ILP to learning patterns for link discovery
“Relational Data Mining with Inductive Logic Programming for Link Discovery” –>