Volume : VIII, Issue : VIII, August - 2018

POISSON FREQUENT PATTERNS CLUSTERING AND TEMPORAL SIMILARITY TRAFFIC MINING FOR WEB USER TRACKING

Ulaganathan. N, Prasath. S

Abstract :

Web traffic pattern analysis is significant to find theweb user behaviors. Few research works has been developed for web traffic pattern mining and web user tracking. However, computational complexity taken for trackinguser location of web traffic patterns was higher. In order to overcome such limitation, Poisson Fragment Frequency based Web Pattern Clustering (PFF–WPC) technique is proposed. The PFF–WPC technique is designed with objective of reducing thecomputational complexity ofidentifying the location of user interest web pages from a weblog database.  Initially PFF–WPC technique performs Poisson Fragment Process with aiming at grouping the web pages in a weblog database according to number of sessions. After session identification, PFF–WPC technique performs Frequency based web patterns clustering with objective of grouping the web pages in an each session as frequent or non frequent web pages with higher clustering accuracy. Finally, PFF–WPC technique carried outs theTemporal Similarity Based Web User Tracking processin whichtraffic web patterns are detected based on the measurement of temporal similarity among the sessions. Based on identified traffic web patterns, the location of corresponding web users is tracked with help of public IP address stored in weblog database with minimum computational complexity. The PFF–WPC technique conducts the experimental evaluations on factors such asclustering efficiency, computational complexity, true positive rate andspace complexity. The experimental result reveals that PFF–WPC technique is able to improve the true positive rate and also reducescomputational complexity of web user tracking when compared to state–of–the–art–works.

Keywords :

Article: Download PDF   DOI : 10.36106/ijar  

Cite This Article:

ULAGANATHAN.N, PRASATH.S, POISSON FREQUENT PATTERNS CLUSTERING AND TEMPORAL SIMILARITY TRAFFIC MINING FOR WEB USER TRACKING, INDIAN JOURNAL OF APPLIED RESEARCH : Volume-8 | Issue-8 | August-2018

Number of Downloads : 729

References :