In this paper we face the coverage problem in the context of learning in the hybrid language Aℒ-log. Here candidate hypotheses are represented as DATALOG clauses with variables constrained by assertions in the description logic AℒC. Regardless of the scope of induction we define coverage relations for Aℒ-log in the two logical settings of learning from implications and learning from interpretations. Also, with reference to the ILP system Aℒ-QUIN, we discuss our solutions to the algorithmic and implementation issues raised by the coverage test for the setting of characteristic induction from interpretations in Aℒ-log.
Efficient Evaluation of Candidate Hypotheses in AL-log
LISI, Francesca Alessandra;ESPOSITO, Floriana
2004-01-01
Abstract
In this paper we face the coverage problem in the context of learning in the hybrid language Aℒ-log. Here candidate hypotheses are represented as DATALOG clauses with variables constrained by assertions in the description logic AℒC. Regardless of the scope of induction we define coverage relations for Aℒ-log in the two logical settings of learning from implications and learning from interpretations. Also, with reference to the ILP system Aℒ-QUIN, we discuss our solutions to the algorithmic and implementation issues raised by the coverage test for the setting of characteristic induction from interpretations in Aℒ-log.File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.