User Characterization and Content Personalization Using Distributed Locality Sensitive Hashing over a Peer-to-Peer Network


Multimedia data are of huge demand from “connected” consumers and how to deliver them effectively and efficiently becomes a new frontier of computer networking research and development. The proposed project uses the research at the University of Victoria on content-based musical information retrieval and the development of Disternet Inc. on a new generation of home gateway devices to enable efficient and effective user characterization and content personalization using distributed locality sensitive hashing over a peer-to-peer network. The research and development is expected to drastically reduce the cost for service providers and network operators to deliver high-quality multimedia contents to end users, and at the same time allows graduate students to use their research to solve real-world problems. 

Aidin Faghfouri
Faculty Supervisor: 
Dr Jianping Pan
British Columbia