摘要
基于内容算法与社会过滤算法都是迄今为止在社交网络中较为成功的好友推荐算法。结合两者的优点,根据用户已有的好友来给用户推荐新的好友,并与用户的兴趣爱好、地理位置等个人信息相结合的方式来处理好友推荐问题。通过实验验证以及准确率和召回率的评测显示,改进的算法比传统的好友推荐算法在推荐性能上有较为明显的提高。
Content -based algorithm and social filtering algorithm are all successful algorithms in social network friend recommendation . This paper combines the advantages of the two algorithms , basing make new friends by means of connecting with users ’ old friends , and combining personal information such as users ’ interests , geographical location etc . to solve the problem of friends recommendation . Through experiments and the precision rate and recall rate evaluation , it showed that the new algorithm is more improved than the traditional friend recommendation algorithm in recommendation functions .
出处
《微型机与应用》
2013年第14期75-78,82,共5页
Microcomputer & Its Applications
关键词
社会过滤
好友推荐
内容相似性
基于内容算法
social filtering
recommendation of friends
content similarity
content-based algorithm