基于人工鱼群算法的协同过滤推荐算法

Collaboration filtering recommendation algorithm based on artificial fish swarm algorithm

基于原始人工鱼群算法,提出在觅食行为中保留较优值以替代随机值,在追尾和聚群行为中比较最优值和中心值再作移动行为的选择,在迭代进行中,实现视野的自适应调整。这样改进后的人工鱼群算法应用于协同过滤推荐系统中,实现用户聚类,从而提高协同过滤推荐系统的最近邻查询速度,降低搜索开销。实验测试结果显示了改进的人工鱼群算法具有收敛速度快,稳定性高的特性,且能获得较优的聚类目标值。将改进的人工鱼群算法用于协同过滤推荐算法中,提高了算法的推荐精度。

Based on the study of the artificial fish swarm algorithm,it is proposed that the preying behavior is improved by keeping the best individual instead of the random individual,the swarming behavior and following behavior are improved by comparing the best individual and the center individual to choice the moving acts,in the iteration,the vision of artificial fish is dynamically adjusted.The improved artificial fish swarm is applied in the collaboration filtering recommendation algorithm,which realize user clustering,for improving the query speed of the nearest neighbor in the collaborative filtering recommendation system,reducing the search spending.The experiment test shows that the improved artificial fish swarm algorithm have some advantages such as faster execution speed,higher stability,and get the optimum clustering,finally it is verified that improved artificial fish swarm is applied in the collaboration filtering recommendation algorithm enhance the precision of recommendation.