Data mining approach for intelligent e services recommendation system using fuzzy rules

Abstract

Recommender systems are nowadays used on a larger scale in e Learning field to populate the items of interest to the online users These users including teachers students and researchers are in need of relevant recommendations lists than those containing mostly irrelevant or unordered recommendations This research work aims at generating a list of recommendation alternatives with highest rankings among the anticipatory ratings of various key concepts ready to be read through by the online users This novel approach makes use of a fuzzy family tree similarity algorithm to select the key concepts that are of more interest to the online users Empirical evaluations prove that the proposed technique is efficient and feasible in including the key concepts in the recommendation list which would newlineotherwise be left out in the conventional tree similarity technique Anticipatory ratings are determined based on the recommendation alternatives User Key Concept Rate UKCR matrix and neighbors sorted in the order of corresponding semantic and content similarities Relevant and suitable content recommendation is an important and challenging task in e Learning In such a scenario, the relevant terms are retrieved for the user in a Recommender System RS which should be able to cope up with varying user preferences over a period of time Next research work proposes a novel recommendation system which provides suitable contents by refining the final frequent item patterns evolving from frequent pattern mining technique and by classifying the final contents using fuzzy logic as simple medium and high level contents This enhancement is achieved by generating frequent item patterns after consolidating the user interest changes with an extended error margin quotient Moreover fuzzy rules are used in this work to enable the rule mining constraints for accommodating all types of learners by applying them on pattern tables Since this method aims in mining the data stream preferences into equal sized windows of user interests it also caters to dy