摘要
为了实现在线推荐信息服务,要对网络号百用户的访问行为进行分析,获取用户访问聚类模型,从而在聚类模型的基础上进行在线推荐。介绍获取用户访问路径信息的方法,对用户访问路径信息建立相似度矩阵,基于相似度矩阵改进K-means算法,据此进行用户模型聚类,给出分析案例,并说明算法实现过程。
In order to realize online information recommendation service, it needs to analyse users' access behavior, get the clustering model for users' access, and realize online recommendation based on this clustering model. Introduces the method to obtain user's access path information, establishes the similarity matrix for the users' access path, improves K-means clustering algorithm based on the similarity matrix, and uses the matrix to cluster user model, provides analysis cases to illustrate the implementation process of algorithm.
出处
《现代计算机》
2010年第4期44-48,共5页
Modern Computer
关键词
路径兴趣度
用户兴趣模型
挖掘
相似度矩阵
Path Interestingness
User Interest Model
Mining
Similarity Matrix