A Novel Web Page Recommender System for Anonymous Users Based on Clustering of Web Pages

  • Rajnikant Wagh
  • Jayantrao Patil

Abstract

Information overload is major problem of
today’s internet use. Users frequently get much more
information than needed. Web Personalization and
recommender systems are becoming popular now days to
overcome this problem. We have proposed a novel web page
recommender system to improve browsing experience of
anonymous users. We have used web usage mining technique
for personalizing a web site and recommendation of web pages.
This technique uses preprocessing, analysis for finding the
relationship among web pages, clustering and classification
phases of data mining. The preprocessing step aims at
maintaining consistency in dataset. We have modelled the
relationship among web pages with novel measures of distance
matrix, occurrence frequency matrix and relationship matrix.
A virtual graph is created corresponding to the relationship
matrix to show the relationship among web pages. The
proposed method partitions the virtual graph into various
clusters i.e. navigation patterns by proposing an enhanced
depth first search algorithm. It is a graph based partitioning
algorithm. We classify the active user under consideration into
one of the cluster by using LCS algorithm. Finally, we used a
threshold value to recommend only optimum number of web
pages. Use of novel measures for finding the relationship, use of
threshold values at the time of formation of clusters as well as
at the time of recommendation of web pages gives us better
results in term of improved visit coherence, accuracy, coverage
and F1 measures. We get max. 61% accuracy, 49.2% avg.
coverage and 28.87% avg. F1 values in the recommendation of
web pages. Similarly, we get 57.8% avg. visit coherence in the
formation of clusters and a minimum of 15 % outliers.

Keywords: Web Personalization, Recommender Systems, Web Usage Mining, Clustering, Classification

References

[1] M. Eirinaki and M. Vazirgiannis, (2003), “Web Mining for Web
Personalization”, In Proceedings of ACM Transactions on Internet
Technology (TOIT).ACM, Athens, Greece, 3(1), pp.1-38,
http://doi.acm.org/10.1145/643477.643478
[2] G. Adomavicius and A. Tuzhilin, (2005), “Toward the Next
Generation of Recommender Systems: A Survey of the State-of-the-
Art and Possible Extensions”, IEEE Transactions on Knowledge and
Data Engineering, 17(6), pp.734–749.
[3] U. Gulden and M. Matthew, (2008), “Personalization Techniques and
Recommender Systems: Series in Machine Perception and Artificial
Intelligence”, World Scientific Press, Vol. 70, Singapore,
[4] M. Bamshed and A. Sarabjot Singh, (2005), “Intelligent Techniques
for Web Personalization”, Springer, New York.
[5] R. Francesco, R. Lior, and S. Bracha, (2015), “Recommender
Systems Handbook”, Springer, New York.
[6] Z. Malik and C. Fyfe, (2012), “Review of Web Personalization”,
Journal of Engineering Technologies in Web Intelligence, 4(3).
[7] Q. Yang, J. Fan, J. Wang, and L. Zhou, (2010), “Personalizing Web
Page Recommendation via Collaborative Filtering and Topic-Aware
Markov Model”, IEEE International Conference on Data Mining,
1(1), pp. 1145-1150.
[8] Y. AlMurtadha, N. Sulaiman, N. Mustapha and N. Udzir, (2011),
“IPACT: Improved Web Page Recommendation System Using Profile
Aggregation Based On Clustering of Transactions”, American Journal
of Applied Sciences, 8 (3), pp. 277-283.
[9] M. Jalali, N. Mustapha, N. Sulaiman and A. Mamat, (2010),
“WebPUM: A Web-based Recommendation System to Predict User
Future Movements”, Expert Systems Applications, 37, pp. 6201-6212.
[10] H. Liu and V. Keselj,(2007), “Combined Mining of Web Server Logs
and Web
[11] Contents for Classifying User Navigation Patterns and Predicting
Users’ Future Request”, Data and Knowledge Engineering, 61(2),
pp.304-330.Conference Short Name:WOODSTOCK’18
[12] B. Mobasher, R. Colley, and J. Shrivastav, (2000), “Automatic
Personalization based on Web Usage Mining”, Communications of
the ACM, 43 (8), pp. 142-151.
[13] C. Sumathi, R. Valli and T. Santahnam,(2010), “An Application of
Session Based Clustering to Analyze Web Pages of User Interest from
Web Log Files”, Journal of Computer Science, 6(1), pp.785-793.
Statistics
0 Views | 0 Downloads
How to Cite
Wagh, R., & Patil, J. (2019). A Novel Web Page Recommender System for Anonymous Users Based on Clustering of Web Pages. Asian Journal For Convergence In Technology (AJCT). Retrieved from http://www.asianssr.org/index.php/ajct/article/view/735
Section
Article