Series in Distributed
edited by Roger Wattenhofer
Understanding and Organizing User Generated Data:
Methods and Applications.
1st edition/1. Aufl. 2011, XII, 164 pages/Seiten, € 64,00.
ISBN 3-86628-381-7 and 978-3-86628-381-7
The Web 2.0, together with the trend towards mobile terminals, has recently had an immense impact on the society. As a result, users nowadays produce data in an unprecedented scale. This new kind of data has attracted the attention of researchers from a wide variety of disciplines. In this thesis we contribute some small pieces to the ongoing efforts to understand this data, thereby focusing on practical aspects accessible to the end-users by means of concrete applications.
In the first part of the thesis we investigate the interconnections between people in (online) social networks, and propose two mobile applications, one to assist group communication, and one in the area of friend finding.
The second part of the thesis studies the extraction of similarity measures from user generated content. We thereby focus on two main domains: Scientific conferences and music. We show that a large collection of publication records implicitly contains information about different aspects of conference similarity. Amongst others, this information is used to implement a conference search engine.
Towards the end of the thesis we discuss different facets of music similarity and its use in end-user applications. In particular, we take advantage of the fact that the cumulated listening histories of a large user basis contain valuable information about the similarity of songs. We then present two different techniques to generate music maps using this similarity information, and finally demonstrate the practical usefulness of the concept of a music map in a comprehensive mobile music player for the Android platform. In a user study, we show that the integrated similarity aware music retrieval interfaces are frequently used and well accepted by the community.
About the author:
Michael Kuhn received his M.Sc. degree in information technologies and electrical engineering from ETH Zurich, Switzerland, in 2004. In 2005 he joined the Distributed Computing Group of Professor Roger Wattenhofer at ETH Zurich as a Ph.D. student and research assistant. In 2010 he earned his Ph.D. degree for his work on the organization of user generated data.
Keywords: user generated data, user interfaces, mobile, social networking, music information retrieval, academic conference search, embedding, applications, recommender systems
Direkt bestellen bei / to order directly from: Hartung.Gorre@t-online.de
Hartung-Gorre Verlag / D-78465 Konstanz / Germany