Sharing Music and Contextual Information in Mobile Social Networks

Tore Nygaard-Jensen

AbstractThis project implements a music and context sharing prototype both with a centralized and decentralized architecture. Information from a set of selected context features is gathered in a user survey with seven users and data analysis is carried out on the collected data to find patterns that can enable creation of new mobile services for the end-users. The collected data are analyzed to discover significant locations from each user and significant social groups are obtained using hierarchical clustering. Music information collected from the users are analyzed to find similarities between the users with collaborative filtering based on implicit ratings to be able to give personalized music recommendations to the users. The music sessions of the users are categorized and indicate in which contexts the users listen to music.
TypeMaster's thesis [Academic thesis]
Year2010    Month May
Publication linkhttp://orbit.dtu.dk/getResource?recordId=260332&objectId=1&versionId=1
BibTeX data [bibtex]
IMM Group(s)Intelligent Signal Processing