Analyzing Communal Tag Relationships for Enhanced Navigation and User Modeling
Simpson, Edwin; Butler, Mark H.
External - Copyright Consideration
Keyword(s): Tagging, folksonomy, clustering, latent structure, navigation, visualization, recommendation, personalization, contextualization
Abstract: The increasing amount of available information has created a demand for better, more automated methods of finding and organizing different types of information resource. This chapter investigates methods of improving navigation, personalization and recommendation of information resources using collaboratively generated tags to model resources and users. We discuss the advantages and limitations of tags, and describe using relationships between tags to discover latent structures that could be used to automatically organize a community's tags. We give a hierarchical clustering algorithm for extracting latent structure and explain methods for determining tag specificity. Next we explain how latent structure visualizations could enhance navigation. Finally we discuss future trends including using latent tag structures to model users and their current tasks for recommendation and user interface personalization. Publication Info: Submitted to (Book) Collaborative & Social Information Retrieval and Access: Techniques for Improved User Modeling, Edited by Max Chevalier, Christine Julien and Chantal Soule-Dupuy, published by IGI Global.
Back to Index