Uncovering Latent Personas in Social Tagging

Managing millions of friends and followers on social sites is increasingly challenging, time consuming, expensive and complicated for a growing number of companies around the world. Effective tools and resources are needed to manage and leverage mission-critical information on people, content and activities in their social networks, blogs, and other social media outlets. the main problem to be tackled in this research is to answer the question “given user personomies and community folksonomies, what are their preferences and interests?� By extracting the information latent in personomies and folksonomies, our goal is to infer the unique insights about users and communities in collaborative tagging sites. Our proposed effort fundamentally differs from other related research in that our focus is on individuals, while others focus on items and content alone. Resolving the complexities described in this proposal will enable our technology to infer people’s connections by monitoring behavior, accurately computing sentiment, uncovering previously inaccessible facets of user persona, and matching Persona Fingerprints across social sites to determine aliases.