期刊名称：International Journal of Computer Science and Information Technologies
电子版ISSN：0975-9646
出版年度：2017
卷号：8
期号：3
页码：443-446
出版社：TechScience Publications
摘要：the Web has undergone a tremendous growthregarding both content and users. This has lead to aninformation overload problem in which people are finding itincreasingly difficult to locate the right information at theright time. Recommender systems have been developed toaddress this problem, by guiding users through the big oceanof information. Until now, recommender systems have beenextensively used within e-commerce and communities whereitems like movies, music and articles are recommended. Morerecently, recommender systems have been deployed in onlinemusic players, recommending music that the users probablywill like.Clustering is an automatic learning technique aimed atgrouping a set of objects into subsets or clusters. The goal is tocreate clusters that are coherent internally, but substantiallydifferent from each other. Automatic document clustering hasplayed an important role in many fields like informationretrieval, data mining, etc. The aim of this thesis is to improvethe efficiency and accuracy of document clustering.All documents and data are in digital form reason of easymaintaining, faster access and compact storage. To accessrelative document easily document clustering is used.Document clustering creates segments collection of textualdocuments into subgroups using similar contents. The purposeof document clustering is to meet human interests ininformation searching and understanding. An effectivefeature phrase of document is more informative feature forimproving document clustering.
