Author : Pushpa C N, Thriveni J, Venugopal K R and L M Patnaik

The exponential explosion of various contents on the Web, made Recommendation Systems increasingly indispensable. Innumerable different kinds of recommendations are made on the Web every day, including movies, music, images, books recommendations, query suggestions and tags recommendations, etc. The paper aims to provide the users with most relevant results (URL’s) to the respective query word. The developed System uses the K-means technique and the Modified System uses the Self-Organizing Map technique. Both the methods use historical browsers data for search key words and provide users with most relevant web pages. All users click-through activity such as number of times he visited, duration he spent, and several other variables are stored in database. The Systems use this database and process to cluster and rank them. The results obtained shows that the Self Organizing Map technique produce most relevant results for a particular query word compared to K-means technique. The Self Organizing Map technique is the optimal method for Web Page recommendations. The Modified System can be utilized in many recommendation tasks on the World Wide Web, including expert finding, image recommendations, image annotations, etc. The experimental results show the promising future of our work.

Keywords: Web Page Recommendations, K-Means Technique, Self-Organizing Map Technique.

Article published in International Journal of Current Engineering and Technology, Vol.4, No.5 (Oct-2014)