International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 78 - Number 8 |
Year of Publication: 2013 |
Authors: Maral Kolahkaj, Ali Haroun Abadi, Mehdi Sadegh Zade |
10.5120/13510-1278 |
Maral Kolahkaj, Ali Haroun Abadi, Mehdi Sadegh Zade . A Recommender System for Web Mining using Neural Network and Fuzzy Algorithm. International Journal of Computer Applications. 78, 8 ( September 2013), 20-24. DOI=10.5120/13510-1278
Web mining is, using data mining tools to discover knowledge from various sources in the web and according to sources type that mined, to be classified in various fields of research. One of the important tools in web mining is mining of web user's behavior that is considered as a way to discover the potential knowledge of web user's interaction. By identifying user's behavior in some cases like: Targeted advertisement, e-commerce, and search engines, it would be possible to provide users with desired results. By providing information that users are interested to view it, users can be converted into permanent customers. In this article, by identifying user's behavior and use of neural and fuzzy techniques it would present a system that will predict user's interest and will propose them a list of pages based on their interests. So that it would enjoy fuzzy clustering method. Due to the user's different interest and use of one or more interest in a time, their use may belong to several clusters. Fuzzy clusters provide a possible overlap. Then by resulting cluster it would extract fuzzy rules. After that, it will make user's movement pattern and with the help of neural network it will propose a list of suggested pages to the users. The results show that the proposed algorithm is in a higher level of precision and recall compared with other algorithm.