Download:
|
by Jean-daniel Zucker, Jean-gabriel Ganascia
In Proc. 8th International Conference on ILP
http://www-apa.lip6.fr/~ganascia/Archives_postcript/ILP98.ps
Add To MetaCart
Abstract:
Abstract. This paper describes a new kind of language bias, S-structural indeterminate clauses, which takes into account the meaning of predicates that play a key role in the complexity of learning in structural domains. Structurally indeterminate clauses capture an important background knowledge in structural domains such as medicine, chemistry or computational linguistics: the specificity of the component/object relation. The REPART algorithm has been specifically developed to learn such clauses. Its efficiency lies in a particular change of representation so as to be able to use propositional learners. Because of the indeterminacy of the searched clauses the propositional learning problem to be solved is a kind of Multiple-Instance problem. Such reformulations may be a general approach for learning non determinate clauses in ILP. This paper presents original results discovered by REPART that exemplify how ILP algorithms may not only scale up efficiently to large relational databases but also discover useful and computationally hard-to-learn patterns. 1.
Citations
|
843
|
Efficient induction of logic programs
– Muggleton, Feng
- 1990
|
|
747
|
Learning logical definitions from relations
– Quinlan
- 1990
|
|
345
|
Inductive logic programming : Theory and methods
– Muggleton, Raedt
- 1994
|
|
152
|
Clausal discovery
– Raedt, Dehaspe
- 1997
|
|
115
|
Solving the multiple-instance problem with axis-parallel rectangles
– Dietterich, Lathrop, et al.
- 1997
|
|
96
|
Controlling the complexity of learning in logic through syntactic and taskoriented models
– Kietz, Wrobel
- 1992
|
|
74
|
Mutagenesis: ILP experiments in a nondeterminate biological domain
– Srinivasan, Muggleton, et al.
- 1994
|
|
71
|
PAC-learnability of determinate logic programs
– Dzeroski, Muggleton, et al.
- 1992
|
|
51
|
Relational clich'es: Constraining constructive induction during relational learning
– Silverstein, Pazzani
- 1991
|
|
21
|
Rapid prototyping of ILP systems using explicit bias
– Cohen
- 1993
|
|
19
|
Learnability of restricted logic programs
– Cohen
- 1993
|
|
18
|
The nature of semantic relations
– Chaffin, Herrmann
- 1988
|
|
18
|
Stochastic propositionalization of non-determinate background knowledge
– Kramer, Helma
- 1998
|
|
13
|
Tractable induction and classification in first order logic via stochastic matching
– Sebag, Rouveirol
- 1997
|
|
11
|
Pac-learning nondeterminate clauses
– Cohen
- 1994
|
|
10
|
CHARADE: A rule System Learning System
– Ganascia
- 1987
|
|
7
|
An application of ILP in a musical database: Learning to compose the two-voice counterpoint
– POMPE, MAKSE
- 1996
|
|
6
|
Integrating multiple learning strategies in first order logics
– Giordana, Neri, et al.
- 1997
|
|
4
|
Mode-directed Inverse Resolution
– Muggleton, Srinivasan
- 1994
|
|
3
|
Changes of Representation for Efficient Learning
– Zucker, Ganascia
- 1996
|
|
1
|
Computational Analysis of Mandarin
– Suen
|
|
1
|
Relational Knowledge Discovery in a Chinese Characters Database
– Zucker, Ganascia, et al.
- 1998
|
|
1
|
A knowledge intensive approach to relational concept learning
– Pazzani, Brunk, et al.
- 1991
|
