Inductive Logic ProgrammingStephen Muggleton Inductive logic programming is a new research area emerging at present. Whilst inheriting various positive characteristics of the parent subjects of logic programming an machine learning, it is hoped that the new area will overcome many of the limitations of its forbears. This book describes the theory, implementations and applications of Inductive Logic Programming. |
From inside the book
Results 1-3 of 48
Page 321
... LINUS . In Section 2 , following the approach used in MIS , we define a specialization refinement operator for each of the systems . This enables the comparison of FOIL and LINUS in ... LINUS 321 3 Refinement Operators for FOIL and LINUS.
... LINUS . In Section 2 , following the approach used in MIS , we define a specialization refinement operator for each of the systems . This enables the comparison of FOIL and LINUS in ... LINUS 321 3 Refinement Operators for FOIL and LINUS.
Page 327
... LINUS to cases 2 and 3 , i.e. , to using equality and background predicates only . In this case , b2 , LINUS < b1 , FOIL and bз , 1 , LINUS < b2 , i , FOIL . This is due to the fact that type information is used to restrict the possible ...
... LINUS to cases 2 and 3 , i.e. , to using equality and background predicates only . In this case , b2 , LINUS < b1 , FOIL and bз , 1 , LINUS < b2 , i , FOIL . This is due to the fact that type information is used to restrict the possible ...
Page 331
... LINUS is then as follows : Beam Search Cost LINUS = O ( BeamWidth × LOC × BFLINUS × N + BFLINUS × N + LOC ) X To conclude , as the number of variables in a clause c grows , BFFOIL ( c ) grows accordingly . This leads to a much larger ...
... LINUS is then as follows : Beam Search Cost LINUS = O ( BeamWidth × LOC × BFLINUS × N + BFLINUS × N + LOC ) X To conclude , as the number of variables in a clause c grows , BFFOIL ( c ) grows accordingly . This leads to a much larger ...
Contents
Inductive Logic Programming | 4 |
A Framework for Inductive Logic Programming | 9 |
3 | 21 |
Copyright | |
44 other sections not shown
Other editions - View all
Common terms and phrases
0-subsumes abstraction operators applied approach arguments arity Artificial Intelligence background knowledge Buntine C₁ C₂ CIGOL CLINT Closed World Assumption complete Computer concept description constrained atoms constraint predicates constraint theory constructed contains defined definite clauses described domain theory e₁ efficient facts Figure finite first-order first-order logic flattening FOCL FOIL formula framework function symbols given GOLEM ground clause head heuristic Horn clauses hypothesis implied inductive learning Inductive Logic Programming inference input instance instantiation integrity constraints intended interpretation inverse resolution learnable learning algorithms Lemma LINUS literals Machine Learning method Morgan Kaufmann Muggleton multi-valued logic Negation as Failure negative examples non-monotonic logic occur oracle PAC-learning polynomial positive examples problem Prolog proof tree relation representation restricted RLGG rules saturation Section set of clauses Shapiro skolemized sort theory sorted atoms sparky specific subset substitution T₁ target Theorem true truncation tuples unit clauses variables