مشخصات پژوهش

عنوان	A New Similarity Measure for mproving Recommender Systems Based on Fuzzy Clustering and Genetic Algorithm
نوع پژوهش	مقاله ارائه شده کنفرانسی
کلیدواژه‌ها	similarity measure, pre-estimation, collaborative filtering, item features,demographic information
چکیده	Recommender systems are widely applied in e-commerce web sites to help customers in finding the items they want.A recommender system should be able to provide users with useful information about the items that might be interesting. Similarity measure is the most important factor in recommender system which isused to compute the user similarity. One can propose a precise similarity measure for improving the recommender system results. The purpose of this paper is to introduce a new similarity measure based on the combination of both users profile and users rating records.The major advantages of the proposed measure comparing with the previous ones is using two different information sources which results in precise results. While the previous ones show the similarity according to user profile or rating.Planning a new similarity measure based on combination of different user information sourcese. g.user profile and rating can overcome sparsity and cold start which are the major problems in recommender systems.The experimental results show that the proposed measure can give satisfactory and high quality recommendations.
پژوهشگران	پرهام مرادی دولت آبادی (نفر دوم)، فرشته کیاست (نفر اول)