PYMK: friend recommendation at myspace
Proceedings of the 2010 ACM SIGMOD International Conference on Management of data
The dynamic competitive recommendation algorithm in social network services
Information Sciences: an International Journal
SoS: um algoritmo para identificar pessoas homófilas em redes sociais com o uso da tradução cultural
Proceedings of the 11th Brazilian Symposium on Human Factors in Computing Systems
Structural and Message Based Private Friend Recommendation
ASONAM '12 Proceedings of the 2012 International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2012)
Who should I add as a "friend"?: a study of friend recommendations using proximity and homophily
Proceedings of the 4th International Workshop on Modeling Social Media
Followee recommendation based on text analysis of micro-blogging activity
Information Systems
Hi-index | 0.00 |
More and more people make friends on the Internet. Most of the community websites generate the friend recommendation lists by search engine. Search engine is not an efficient mechanism because the database is too huge that search engine produces many unnecessary and unordered friend lists. However, there is little research discussing the issue of the recommendation quality of friend on the Internet. In this study, we propose a new recommendation algorithm named weighted minimum-message ratio (WMR) which generates a limited, ordered and personalized friend lists by the real message interaction number among web members. We chose 30 potential members from a community website as our experimental cases in this study. The result shows that the best recommended friend number for a target member is 15 and the precision and recall are 15% and 8% for testing prediction, respectively. This result is acceptable compared with book recommendation in which the testing precision and recall are 3% and 14%, respectively.