Title: Frequent pattern sub-space clustering optimisation algorithm for data mining from large database
Authors: T. Sheik Yousuf; M. Indra Devi
Addresses: Department of Computer Science Engineering, Mohammed Sathak Engineering College, Kilakarai, Ramanathapuram, TamilNadu, 623-503, India ' Department of Computer Science Engineering, Kamaraj College of Engineering and Technology, Virudhunagar, TamilNadu, India
Abstract: Data mining environment gives a quick response to the user by fast and correctly pick-out the item from the large database is a very challenging task. Previously, multiple algorithms were proposed to identify the frequent item since they are scanning database at multiple times. To overcome those problems, we proposed Rehashing based Apriori Technique in which hashing technology is used to store the data in horizontal and vertical formats. Rehash Based Apriori uses hashing function to reduce the size of candidate item set and scanning of database, eliminate non-frequent items and avoid hash collision. After finding frequent item sets, perform level wise subspace clustering. We instigate generalised self organised tree based (GSTB) mechanism to adaptively selecting root to construct the path from the cluster head to neighbours when constructing the tree. Our experimental results show that our proposed mechanisms reduce the computational time of overall process.
Keywords: sub-space clustering; generalised self-organised tree-based cluster head selection; GSTB.
International Journal of Business Intelligence and Data Mining, 2018 Vol.13 No.1/2/3, pp.221 - 246
Received: 17 Nov 2016
Accepted: 22 Feb 2017
Published online: 03 Nov 2017 *