Improving the Performance of PSO based Clustering by using Surrogates

Conference: Recent Trends in Information Processing, Computing, Electrical and Electronics
Author(s): Priyanka Shrivastava, Mangesh Khandelwal, Vinod Kumar Year: 2017
Grenze ID: 02.IPCEE.2017.1.1 Page: 1-6

Abstract

Clustering is the grouping of a specific set of objects based on their characteristics, aggregating them according to\ntheir similarities. Regarding to data mining, this methodology partitions the data implementing a specific combine algorithm,\nmost appropriate for the desired information analysis. K-means is among the simplest unsupervised learning algorithms that\nresolve the well-known clustering problem. The procedure follows a simple and effortless way to classify a given data set\nthrough a certain number of clusters. However the method can be computationally costly in that a high number of function\ncalls have to progress the swarm at each optimization iteration. In order to increase the efficiency of this algorithm an notion\nof Surrogate purpose is incorporated which functions as an stand in for pricey objective function. The work also aims to\nprovide better evaluation of the proposed hybrid approach on the basis of acquired numerical results.

<< BACK

IPCEE - 2017