Mining differential dependencies: A subspace clustering approach

Date

2014

Authors

Kwashie, S.
Liu, J.
Li, J.
Ye, F.

Editors

Wang, H.
Sharaf, M.A.

Advisors

Journal Title

Journal ISSN

Volume Title

Type:

Conference paper

Citation

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2014 / Wang, H., Sharaf, M.A. (ed./s), vol.8506 LNCS, pp.50-61

Statement of Responsibility

Conference Name

25th Australasian Database Conference (ADC) (14 Jul 2014 - 16 Jul 2014 : AUSTRALIA, Univ Queensland, Brisbane)

Abstract

The discovery of differential dependencies (DDs) is the problem of finding a minimal cover set of DDs that hold in a given relation. This paper proposes a novel subspace-clustering-based approach to mine DDs that exist in a given relation. We study and reveal a link between δ-nClusters and differential functions (DFs). Based on this relationship, we adopt and co-opt techniques for mining δ-nClusters to find the set of candidate antecedent DFs of DDs efficiently, based on a user-specified distance threshold. Furthermore, we define an interestingness measure for DDs to aid the discovery of essential DDs and avoid the mining of an extremely large set. Finally, we demonstrate the scalability and efficiency of our solution through experiments on real-world benchmark datasets.

School/Discipline

Dissertation Note

Provenance

Description

Access Status

Rights

Copyright 2014 Springer International Publishing Switzerland

License

Grant ID

Call number

Persistent link to this record