Title Of Paper:
 
 
Predicting Web Page Accesses, using Users’ Profile and Markov Models
Author's Name :  Zeynab Fazelipour
KeyWords:  Web Server Log File, User’s Profile, Web Mining, Markov Models
Pages:  10 -19
Volume: 4
Issue: 1
Year: 2015
Abstract:
 

Nowadays web is an important source for information retrieval, the sources on WWW are constantly increasing and the users accessing the web have different backgrounds. Consequently, finding the information which satisfies the personal users needs is not so easy. Exploration of users behaviors in the web, as a method for extracting the knowledge lying behind the way of how the users interact with the web, is considered as an important tool in the field of web mining. By identifying user's behavior in such cases as targeted advertisement, e-commerce, and search engines, it would be possible to provide users with desired results. Web page prediction gained its importance from the ever increasing number of e-commerce Web information systems and e-businesses. Pattern discovery techniques as such helps predict the next page to be accessed a Web user based on the user's previous browsing behavior. In this paper, we present a method to predict the web page where in the profile of users and Markov models are used.User's profile is obtained based on the characteristics of web server log file. Then, applying Markov models on the users sessions with same interest, the sequential patterns are extracted. In the proposed method, after detecting the movement pattern, the user's desired pageis predicted using the relevant Markov model. The evaluation results demonstrate that the proposed algorithm has higher precision comparing to other prediction systems.

Full Text: