TechLens: Enhanced Digital Libraries through Recommendation
The University's Department of Computer Science and Engineering (GroupLens research group), and the University Libraries have partnered to research the use of citations, personal bibliographies, and metadata to synthesize recommender systems library services users. To date, our research has focused on the resolvability of bibliographic references collected by researchers, the incentives and risks relating to the disclosure personally-collected citation data, privacy implications of recommendation systems, and prototype design for a tune-able recommender system that uses citation data.
TechLens is supported in part by the National Science Foundation, and works in partnership with Association for Computing Machinery (ACM) and RefWorks.
Recent TechLens work includes:
TechLens - A Researcher's Desktop (prototype)
http://techlens.cs.umn.edu/
A preliminary prototype that showcases how we can learn about users, and their research interests through their personal collections, And, use that information to offer them recommender based services. Also allows 'ratings' and 'annotations' for a citation. Currently uses the dataset from the Association for Computing Machinery.
Research papers and presentations:
International Conference on Conceptions of Library and Information Science (CoLIS) 2007
"Resolvability of references in users' personal collections" Kapoor, N. et. al.
http://www-users.cs.umn.edu/~nkapoor/pubs/nkapoor_colis07.pdf
Explains resolvability of references in (RefWorks) users' personal online collections.
European Conference on Digital Libraries (ECDL) 2007
"A Study of Citations in Users' Online Personal Collections", Nishikant Kapoor, et. al.
http://www-users.cs.umn.edu/~nkapoor/pubs/nkapoor_ecdl07.pdf
Further discusses the nature of citation data in (RefWorks) users' personal online collections, broken down in users' discipline and educational status. Also talks about overlap between different collections and users' privacy concerns.