摘要
针对现有协同过滤算法具有的可扩展性较低、数据稀疏和计算量较大缺点,提出一种基于SVD矩阵分解技术和RkNN算法的协同过滤推荐算法.本算法经SVD矩阵简化处理和kNN和RkNN的协作过滤,增强了用户的影响集,实现了测试集的未知预测评分功能.经仿真实验表明,稀疏性、可扩展性和计算量都得到有效改善,系统预测评分与用户实际评分接近,为用户提供了良好的使用体验.该算法获得了更好的预测性能,同时具有良好的可扩展性.
The RBF neural network has the disadvantages of slow convergence speed and leak global searching ability, this paper presents a RBF neural network based on genetic algorithm, improves the prediction accuracy of RBF neural network prediction model and realizes nonlinear time sequence after adapting genetic operator parameters optimization. The simulation experiment results show that the prediction of RBF network prediction model is very suitable for the genetic algorithm based on nonlinear time series. It is feasible, accurate and effective.
出处
《湖南工程学院学报(自然科学版)》
2015年第1期44-47,共4页
Journal of Hunan Institute of Engineering(Natural Science Edition)