In this paper we investigate different methods of selecting seed examples and their effect on learning first order rules in ILP. Based on large experimental results we concluded that they have significant impact on accuracies, the number of rules and the speed of induced models.
|Translated title of the contribution||Selection of Seed Examples in ILP|
|Title of host publication||The 3rd York Doctoral Symposium on Computing (YDS'10)|
|Publisher||Department of Computer Science, The University of York|
|Publication status||Published - 2010|
Bibliographical noteOther page information: 29-38
Conference Proceedings/Title of Journal: The 3rd York Doctoral Symposium on Computing (YDS'10)
Other identifier: 2001271