作者: Diane J. Cook , Lawrence B. Holder , Jeff Coble , Joseph Potts
关键词: Cluster analysis 、 Graph (abstract data type) 、 Graph database 、 Inductive logic programming 、 Minimum description length 、 Theoretical computer science 、 Supervised learning 、 Molecule mining 、 Relational database 、 Computer science
摘要: We describe an approach to learning patterns in relational data represented as a graph. The approach, implemented the Subdue system, searches for that maximally compress input can be used supervised learning, well unsupervised pattern discovery and clustering.