Title :
An Online Bayesian Networks Model for E-Commercial Personalized Recommendation System
Author :
Zhang, Shao-Zhong ; Liu, Lu ; Dong, Yu-Zhi
Abstract :
This paper presents an online personalized recommended model based Bayesian networks. The model adopts outline learning algorithm for structure and online learning algorithm for parameter. The online parameter learning realizes the online adjusting and correcting for model. The online algorithm is based on the theory of EM and introduces correcting functions to realize online EM. The experimentation shows that the algorithm in this paper is suitable for online learning of personalized recommended model and the model that is obtained by online EM has a higher precision than basic EM.
Keywords :
Bayes methods; electronic commerce; expectation-maximisation algorithm; information filters; learning (artificial intelligence); e-commerce; e-commercial personalized recommendation system; expectation-maximisation algorithm; online Bayesian networks model; online parameter learning; Bayesian methods; Conference management; Cybernetics; Electronic mail; Filtration; Machine learning; Management information systems; Merchandise; Microelectronics; Transaction databases; Bayesian networks; E-commercial recommended systems; Online model;
Conference_Titel :
Machine Learning and Cybernetics, 2007 International Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-0973-0
Electronic_ISBN :
978-1-4244-0973-0
DOI :
10.1109/ICMLC.2007.4370754
