Chapter

Advanced Data Mining and Applications

Volume 4632 of the series Lecture Notes in Computer Science pp 323-334

AOG-ags Algorithms and Applications

  • Lizhen WangAffiliated withCarnegie Mellon UniversityDepartment of Computer Science and Engineering, School of Information, Yunnan, University, Kunming, 650091Department of Informatics, School of Computing and Engineering, University of, Huddersfield, Huddersfield, HD1 3DH
  • , Junli LuAffiliated withCarnegie Mellon UniversityDepartment of Computer Science and Engineering, School of Information, Yunnan, University, Kunming, 650091
  • , Joan LuAffiliated withCarnegie Mellon UniversityDepartment of Informatics, School of Computing and Engineering, University of, Huddersfield, Huddersfield, HD1 3DH
  • , Jim YipAffiliated withCarnegie Mellon UniversityDepartment of Informatics, School of Computing and Engineering, University of, Huddersfield, Huddersfield, HD1 3DH

* Final gross prices may vary according to local VAT.

Get Access

Abstract

The attribute-oriented generalization (AOG for short) method is one of the most important data mining methods. In this paper, a reasonable approach of AOG (AOG-ags, attribute-oriented generalization based on attributes’ generalization sequence), which expands the traditional AOG method efficiently, is proposed. By introducing equivalence partition trees, an optimization algorithm of the AOG-ags is devised. Defining interestingness of attributes’ generalization sequences, the selection problem of attributes’ generalization sequences is solved. Extensive experimental results show that the AOG-ags are useful and efficient. Particularly, by using the AOG-ags algorithm in a plant distributing dataset, some distributing rules for the species of plants in an area are found interesting.

Keywords

Attribute-oriented generalization (AOG) Concept hierarchy trees Attributes’ generalization sequences (AGS) Equivalence partition trees Interestingness of AGS