Socialtagger - collaborative tagging for blogs in the long tail

Abstract

Social bookmarking is the process through which users share tags for online resources like blogs with others. Such collaborative tags provide valuable metadata for retrieval systems. While the successes of collaborative tagging systems have been demonstrated by popular websites like Del.icio.us, these sites cover only a small fraction of the available blogs on the web. The vast majority of the blogs are not available on any collaborative tagging system and are often tagged only by the authors. This lack of coverage of collaborative tags is a considerable roadblock in using the tag metadata in a web scale information retrieval system. To solve this problem we propose and implement a system to automatically recommend collaborative tags for a blog. The automatically generated tags will help to surface the blogs by making them available on social book marking sites and allow them to be easily discovered and potentially further tagged by a wider population.

On social networks and collaborative recommendation

“…Subramanya and Liu [17] propose a system that automatically recommends tags for blogs, using similarity ranking in a manner similar to collaborative filtering techniques. Stromhaier [16] studies a novel idea in tag recommendation, which bridges the gap between the keywords issued by a user in a query and the tags actually used by a social system.…”

Uses of explicit and implicit tags in social bookmarking

“…Tags can also be considered as metadata (Subramanya & Liu, 2008), i.e., data about data. The three types of metadata are administrative, structural, and descriptive (Taylor, 2003) and can be developed by dedicated professionals, authors, or general users (Mathes, 2004).…”

Sign-Based Semantics-Enriched Social Computing

“…Many similar semantic techniques can be found. For example, SocialTagger [24] used the multivariate statistical technique of canonical correlation analysis to associate blogs and tags and thus find their semantic relations. Other examples employing associations and similarities could be [13][19] [26].…”