Self-adaptive recommendation policy in recommender systems
Donglin Chen · Computer Engineering and Applications Journal · 2007
A method of choosing recommendation policy adaptive to the recommendation context is proposed in this paper to overcome the shortcomings of different recommendation techniques.The recommendation context is denoted as a duple user knowledge,commodity category.The clusters of the recommendation context are acquired by exploiting the ART artificial neural network.The recommendation technique of best recommendation quality corresponding to each cluster is acquired by the analysis of reflection from users on the recommendation results.So the recommendation technique of best quality can be applied to generate recommendation results for user in correspondence to the cluster in which the duple user knowledge,commodity category of the user reside in.The working process of the system is self-adaptive without manual intervention.