[an error occurred while processing this directive]
���¿��ټ��� �߼�����
   ��ҳ  �ڿ�����  ��ί��  Ͷ��ָ��  �ڿ�����  ��������  �� �� ��  ��ϵ����
�������պ����ѧѧ�� 2009, Vol. 35 Issue (9) :1057-1061    DOI:
���� ����Ŀ¼ | ����Ŀ¼ | ������� | �߼����� << | >>
�� Ʒ, �Ѿջ�, ������, �� �*
�������պ����ѧ �����ѧԺ, ���� 100191
Learning drifting user interest incrementally from numerically labeled feedbacks
Zhang Pin, Pu Juhua, Liu Yongli, Xiong Zhang*
School of Computer Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100191, China

Download: PDF (0KB)   HTML 1KB   Export: BibTeX or EndNote (RIS)      Supporting Info
���µ��� [1]��Koychev I, Schwab I. Adaptation to drifting users interest ECML Workshop:Machine Learning in New Information Age. Barcelona, Spain:[s.n.], 2000:39-45�� [2]��Billsus D, Pazzani M J. A hybrid user model for news classificationThe Seventh International Conference on User Modeling. Wein, New York: Springer��Verlag, 1999:99-108�� [3]��Forman G. Tackling concept drift by temporal inductive transferThe 29th Annual International ACM SIGIR Conference. Seattle, Washington: Springer��Verlag,2006:252-259�� [4]��Widyantoro D H, Ioerger T R, Yen J. Learning user interest dynamics with a three��descriptor representation[J]. J Am Soc Information Science, 2001, 52(3):212-225�� [5]��Widmer G, Kubat M. Learning in the presence of concept drift and hidden contexts [J]. Machine Learning, 1996, 23(1): 69-101�� [6]��Kearns M J, Schapire R E. Efficient distribution��free learning of probabilistic concepts31st Annual Symposium on Foundations of Computer Science. St Louis, MO: IEEE Computer Society Press, 1990(1): 382-391�� [7]��Ji J, Liu C, Yan J, et al. Bayesian networks dtructure learning and its application to personalized recommendation in a B2C portalInternational Conference on Web Intelligence. Beijing: Computer Society, 2004:179-184�� [8]��Kamahara J, Asakawa T, Shimojo S, et al. A community��based recommendation system to reveal unexpected interests11th International Multimedia Modelling Conference. Melbourne, Australia: IEEE Computer Society, 2005:433-438[9]��Good N, Schafer J, Konstan J, et al. Combining collaborative filtering with personal agents for better recommendationsSixteenth National Conferenceon on Artificial Intelligence. Orlando, Florida: MIT Press, 1999: 439-446  [10]��Singh S, Shepherd M, Duffy J, et al. An adaptive user profile for filtering news based on a user interest hierarchy69th Annual Meeting of the American Society for Information Science and Technology. Austin, Texas: Wiley, 2006(43):1-27�� [11]��Kim C O, Kwon I H, Baek J G.. Asynchronous action��reward learning for nonstationary serial supply chain inventory control [J]. Applied Intelligence, 2008, 28(1):1-16�� [12]��Mika S, Jain L C. Introduction to fuzzy clustering [J]. Innovations in Fuzzy Clustering, 2006,205:1-8�� [13]��Hendratmo W D. Concept drift learning and its application to adaptive information filtering[D]. Texas: Texas A&M University, 2003�� [14]��Hofmann T. Collaborative filtering via gaussian probabilistic latent semantic analysis26th Annual International ACM SIGIR. New York: Association for Computing Machinery, 2003:259-266
ժҪ ���������ȤƯ������ѧϰ���������԰�����ֵ���ݱ�ǩ���û���������ѧϰ�IJ���,�����һ��������������ֵ��ǩ��������ȤƯ������ѧϰ����,�Ի�Ȼ�������ʽ�����û���Ȥ,���û������а��������ݱ�ǩ��Ϊ�û��Ը�ʵ����ϲ�ø���,�����û���ָ�����ȼ�Ȩƽ���ķ�������Ȥģ�ͽ�������ѧϰ.�ڲ�ͬѧϰ�����µ�ʵ��������,�÷����ܹ��ڷ������ݱ�ǩΪ������ֵ�������´ﵽ�����з������õ�ѧϰЧ��.
Email Alert
�ؼ����� ѧϰ�㷨   ǿ��ѧϰ   ģ����     
Abstract�� Most incremental approaches for learning drifting user interests assume that data instances in user feedbacks are binary labeled. A novel incremental learning approach was presented which learns drifting user interests from numerically labeled feedbacks instead of binary labeled ones. User interests were modeled as a set of probabilistic concepts. Numerical instance labels were considered as probabilities that the user likes those instances. Feedbacks were used to update user interest models incrementally based on an exponential, recency-weighted average algorithm. Experimental results on different learning tasks showed that the approach outperforms existing approaches in numerically labeled feedback environment.
Keywords�� learning algorithms   reinforcement learning   fuzzy sets     
Received 2008-09-18;


About author: �� Ʒ(1983-),��,���ɹŰ�ͷ��,˶ʿ��,zhangpin@cse.buaa.edu.cn.
�� Ʒ, �Ѿջ�, ������, �� �.������������ֵ��ǩ����ȤƯ������ѧϰ����[J]  �������պ����ѧѧ��, 2009,V35(9): 1057-1061
Zhang Pin, Pu Juhua, Liu Yongli, Xiong Zhang.Learning drifting user interest incrementally from numerically labeled feedbacks[J]  JOURNAL OF BEIJING UNIVERSITY OF AERONAUTICS AND A, 2009,V35(9): 1057-1061
http://bhxb.buaa.edu.cn//CN/     ��     http://bhxb.buaa.edu.cn//CN/Y2009/V35/I9/1057
Copyright 2010 by �������պ����ѧѧ��