Title
Learning of Class Descriptions from Class Discriminations: A Hybrid Approach for Relational Objects
Abstract
The paper addresses the question how learning class discrimination and learning characteristic class descriptions can be related in relational learning. We present the approach TRITOP/MATCHBOX combining the relational decision tree algorithm TRITOP with the connectionist approach MATCHBOX. TRITOP constructs efficiently a relational decision tree for the fast discrimination of classes of relational descriptions, while MATCHBOX is used for constructing class prototypes.Although TRITOP's decision trees perform very well in the classification task, they are difficult to understand and to explain. In order to overcome this disadvantage of decision trees in general, in a second step the decision tree is supplemented by prototypes. Prototypes are generalised graphtheoretic descriptions of common substructures of those subclasses of the training set that are defined by the leaves of the decision tree. Such prototypes give a comprehensive and understandable description of the subclasses. In the prototype construction, the connectionist approach MATCHBOX is used to perform fast graph matching and graph generalisation, which are originally NP-complete tasks.
Year
DOI
Venue
2002
10.1007/3-540-45751-8_13
KI
Keywords
Field
DocType
hybrid approach,class discriminations,class descriptions,class discrimination,class prototype,relational learning,characteristic class description,relational decision tree,relational description,decision tree,relational decision tree algorithm,approach tritop,relational objects,connectionist approach,characteristic class
Inductive logic programming,Decision tree,Tree (graph theory),Computer science,Statistical relational learning,Matching (graph theory),Artificial intelligence,Class discrimination,Decision tree learning,Machine learning,Connectionism
Conference
Volume
ISSN
ISBN
2479
0302-9743
3-540-44185-9
Citations 
PageRank 
References 
1
0.35
18
Authors
3
Name
Order
Citations
PageRank
Peter Geibel128626.62
Kristina Schädler2373.92
Fritz Wysotzki345645.46