研究、探讨了协同推荐问题,提出了一种基于两层面的多个后向传播(BP)神经网络的协作过滤推荐算法(TMNN-CFRA).两层面的多个BP神经网络协同工作,高层面BP网反向误差传播直至低层面多个人工神经网络(ANN)进行网络权值修正,以此为基础,借助用户评价等特征前向给出项目推荐.标准评测集Movielens上的实验评测表明了TMNN-CFRA的可行性和有效性.
A novel two-level multiple neural networks-based collaborative filtering recommendation algorithm (TMNN-CFA) for rating prediction is presented. By cooperating the multiple back propaga- tion (BP) networks together, the higher layer neural network propagates conversely the output deviation until to the lower layer neural networks to amend the network weights and based on which, item recommendation is accomplished in the forward process of two layers networks relying on the factors such as ratings, etc. Experiment results on the standard Movielens show that TMNN-CFA method is effective and feasible for item recommendation.