Academic Journals Database
Disseminating quality controlled scientific knowledge

Web Usage Data Clustering Using Improved Genetic Fuzzy C-Means Algorithm

Author(s): Karunesh Gupta | Manish Shrivastava

Journal: International Journal of Advanced Computer Research
ISSN 2249-7277

Volume: 2;
Issue: 2;
Start page: 77;
Date: 2012;
VIEW PDF   PDF DOWNLOAD PDF   Download PDF Original page

Keywords: Web Usage Mining | Genetic Algorithm | Fuzzy C-Means

Web usage mining involves application of datamining techniques to discover usage patterns fromthe web data. Clustering is one of the importantfunctions in web usage mining. Recent attemptshave adapted the C-means clustering algorithm aswell as genetic algorithms to find sets of clusters .Inthis paper; we have proposed a new framework toimprove the web sessions’ cluster quality from fuzzyc-means clustering using Improved GeneticAlgorithm (GA). Initially a fuzzy c-means algorithmis used to cluster the user sessions. The refinedinitial starting condition allows the iterativealgorithm to converge to a “better” local minimum.And in the second step, we have proposed a new GAbased refinement algorithm to improve the clusterquality. The proposed algorithm is tested with webaccess logs collected from the UCI datasetrepository.
Affiliate Program      Why do you need a reservation system?