Chitta, Radha and Murty, M Narasimha (2010) Two-level k-means clustering algorithm for k-tau relationship establishment and linear-time classification. In: Pattern Recognition, 43 (3). pp. 796-804.
26.pdf - Published Version
Restricted to Registered users only
Download (407Kb) | Request a copy
Partitional clustering algorithms, which partition the dataset into a pre-defined number of clusters, can be broadly classified into two types: algorithms which explicitly take the number of clusters as input and algorithms that take the expected size of a cluster as input. In this paper, we propose a variant of the k-means algorithm and prove that it is more efficient than standard k-means algorithms. An important contribution of this paper is the establishment of a relation between the number of clusters and the size of the clusters in a dataset through the analysis of our algorithm. We also demonstrate that the integration of this algorithm as a pre-processing step in classification algorithms reduces their running-time complexity.
|Item Type:||Journal Article|
|Additional Information:||Copyright for this article belongs to Elsevier science.|
|Keywords:||Clustering; k-Means; Classification; Linear-time complexity; Support vector machines; k-Nearest neighbor classifier|
|Department/Centre:||Division of Electrical Sciences > Electrical Engineering
Division of Electrical Sciences > Computer Science & Automation (Formerly, School of Automation)
|Date Deposited:||19 Jan 2010 08:57|
|Last Modified:||19 Sep 2010 05:54|
Actions (login required)