论文部分内容阅读
粒子群优化(PSO)模仿鸟群飞行觅食行为,通过粒子追随自己找到的最好解和整个群的最好解来完成优化。信息推荐服务是数字图书馆的一项重要的功能,本文提出应用多目标粒子群优化算法对用户和项之间的相似性同时进行聚类,为用户提供最优的信息推荐服务。在MovieLens数据集的实验结果表明我们的方法能够为用户提供有用的推荐意见,其性能优于其他推荐系统方法。
Particle swarm optimization (PSO) mimics the foraging behavior of flocks and completes the optimization by following the best solution that the particle follows and the best solution of the entire swarm. The information recommendation service is an important function of the digital library. This paper proposes to apply the multi-objective particle swarm optimization algorithm to cluster the similarities between users and items at the same time, and provide users with the best information recommendation service. The experimental results in the MovieLens dataset show that our method can provide users with useful recommendations, and its performance is superior to other recommended system methods.