Analyzing Communal Tag Relationships for Enhanced Navigation and User Modeling

Analyzing Communal Tag Relationships for Enhanced Navigation and User Modeling

Edwin Simpson (HP Labs, UK) and Mark H. Butler (HP Labs, UK)
DOI: 10.4018/978-1-60566-306-7.ch003
OnDemand PDF Download:
$37.50

Abstract

The increasing amount of available information has created a demand for better, more automated methods of finding and organizing different types of information resources. This chapter investigates methods for enabling improved navigation, user modeling, and personalization using collaboratively generated tags. The authors discuss the advantages and limitations of tags, and describe how relationships between tags can be used to discover latent structures that can automatically organize a collection of tags owned by a community. They give a hierarchical clustering algorithm for extracting latent structure and explain methods for determining tag specificity, then use visualization to examine latent structures. Finally the authors discuss future trends including using latent tag structures to create user models.
Chapter Preview
Top

Background

The Web is notable for its increasing abundance of information and the freedom it allows for new innovation, particularly the development of new methods to connect users with information and allow them to organize resources. These methods augment more established approaches such as free-text search (Gudivada et al., 1997; Page et al., 1998), directories and taxonomies (Garshol, 2004), explicit communication between users and subscriptions to news feeds or mailing lists. In this section we describe folksonomies and consider both the advantages and limitations of this approach. We then compare tagging with thesauri and controlled vocabularies, and consider how these techniques can be combined with tagging.

Complete Chapter List

Search this Book:
Reset