Abstract : The recommendation system plays a vital role in web usage mining. It recommends the selected URLs to the users of the system on the basis of various factors with less complexity. A large number of recommendation systems have been developed in literature and few issues observed in those systems are usage of small number of URLs, inefficient factors for classification and recommendation etc. This study develops a novelrecommendation system which performs automatic recommendations for selected keywords using fuzzy inference engine and also addresses the important factors with hyperlinks such as inbound links, outbound links, tags etc. The proposed classification process does not get affected even if the data goes out of bound, as it works not only on distance but also with keywords, matching, tags etc. as major factors. © 2018 IEEE.