Paper Info
Title
Disjunctive Bottom Set and Its Computation
Abstract
This paper presents the concept of the disjunctive bottom set and discusses its computation. The disjunctive bottom set differs from existing extensions of the bottom set, such as kernel sets(Ray, Broda, & Russo 2003), by being the weak- est minimal single hypothesis for the whole hypothesis space. The disjunctive bottom set may be characterized in terms of minimal models. Therefore, as minimal models can be com- puted in polynomial space complexity, so can the disjunctive bottom set. We outline a flexible inductive logic program- ming framework based on the disjunctive bottom set. Com- pared with existing systems based on bottom set, such as Pro- gol (Muggleton 1995), it can probe an enlarged hypothesis space without increasing space complexity. Another novelty of the framework is that it provides an avenue, via hypoth- esis selection function, for the integration of more advanced hypothesis selection mechanisms.
Year
Venue
Keywords
2007
FLAIRS Conference
space complexity
Field
DocType
Citations 
Inductive logic programming,Kernel (linear algebra),PROGOL,Minimal models,Computer science,Algorithm,PSPACE,Novelty,Computation
Conference
0
PageRank 
References 
Authors
0.34
11
2
Name
Order
Citations
PageRank
Wenjin Lu18410.79
Ross D. King21774194.85