Improving collaborative tag recommendation by using local lexicon in social comment context

Folksonomies enable Internet users to share, annotate and search for online resources with tags. Tag recommendation can suggest tags that maximize utility and help to promote resource sharing and collaboration. To date, little has been done to collaboratively recommend tags in narrow folksonomies. In this paper we apply K-Nearest Neighbor to find similar users with common interests who lie close to each other in social networks. We present a novel tag recommendation strategy in social comment context. The evaluation shows that tag recommendation based on local lexicon is an effective way to improve collaborative resource sharing in social networks.