Paper Info
Title
Using SOM to clustering of web sessions extracted by techniques of web usage mining
Abstract
Everyday a huge amount of pages are published on the Web, and, as a consequence, the users' difficulty to locate those that will meet their needs is increasingly bigger. The challenge for web designers and e-commerce companies is to identify groups of users that present similar interests in order to personalize navigation environments to meet those interests. In an attempt to offer that to the countless web users, in the last years, several researches have been done on clustering applied to Web Usage Mining. In this paper, a log file is preprocessed to map the sequence of visits for each user's session. A Session-Path Matrix is used as input to SOM Map and identifying patterns between each session. The results show the similarities between the sessions based on time spent on visited paths and volume transferred.
Year
DOI
Venue
2012
10.1007/978-3-642-32639-4_59
IDEAL
Keywords
Field
DocType
last year,countless web user,e-commerce company,web usage mining,session-path matrix,som map,web designer,huge amount,navigation environment,web session,log file
Web design,World Wide Web,Web mining,Web mapping,Computer science,Web navigation,Web service,Cluster analysis
Conference
Citations 
PageRank 
References 
0
0.34
6
Authors
2
Name
Order
Citations
PageRank
Fábio A. Procópio de Paiva122.06
José Alfredo F. Costa25210.11