Ben Fields

Contents 1 Contact Information 2 Universities 2.1 PhD 2.2 MSc 2.3 BS 3 Current Activities 3.1 Research Interests 3.2 Publications

Contact Information

email: b.fields(at)gold(dot)ac(dot)uk
(mobile) phone: +44 (0) 79 6106 1568

office address:
Ben Fields
Department of Computing
Goldsmiths College
University of London
New Cross
London, SE14 6NW
United Kingdom

Universities

PhD

PhD Computing, expected June 2010
Goldsmiths College, University of London
Proposed title - Contextualized Music Recommendation: The Playlist as Recommendation Engine
My Master's of Philosophy to PhD upgrade document which is something of an extended (30ish pages) proposal and lit review on the topic.

MSc

M.S. Music Engineering Technology, May 2006

University of Miami

Thesis: ON THE VIABILITY OF USING MIXED FEATURE EXTRACTION WITH MULTIPLE STATISTICAL MODELS TO CATEGORIZE MUSIC BY GENRE

available for download: pdf

BS

B.S. Computer Engineering, March 2004
University of California, Santa Cruz
Senior Design Project: Securio - The Autonomous Home Security Robot, Audio Video Specialist

Current Activities

I started my PhD research in music similarity and organization as part of OMRAS2 in January. I am currently working in the Goldsmiths Digital Studios research group, under the supervision of Dr. Micheal Casey.

Research Interests

• bridging the semantic gap between social networks and content based retrieval systems
• Music recommenders
• Metadata informed dynamics effects processing
• Human-like Playlist generation and playback (autoDJ)
• Song phrase segmentation
• Playlist (think DJ set, pod cast, etc.) segmentation

Publications

• B. Fields, K. Jacobson, C. Rhodes, and M. Casey, "Social playlists and bottleneck measurements: Exploiting musician social graphs using content-based dissimilarity and pairwise maximum flow values," in Proc. of Int. Symposium on Music Information Retrieval, September 2008.
• K. Jacobson, B. Fields, and M. Sandler, "Using audio analysis and network structure to identify communities of on-line social networks of artists," in Proc. of Int. Conference on Music Information Retrieval, Philadelphia, PA, USA, October 2008.
• B. Fields, K. Jacobson, M. Casey, and M. Sandler, "Do you sound like your friends? exploring artist similarity via artist social network relationships and audio signal processing," in Int. Computer Music Conference, August 2008.
• B. Fields, M. Casey, "Using Audio Classifiers as a Mechanism for Content Based Song Similarity," in Proc. Audio Engineering Society 123 Int. Conv., (New York, NY, USA), Audio Engineering Society, October 2007

• M. Mauch, S. Dixon, C. Harte, B. Fields, M. Casey, "Discovering Chord Idioms through Beatles and Real Book Songs," in Proc. Int. Symposium on Music Information Retrieval, Vienna, Austria, Sept. 2007.
• B.Fields, “Using mixed feature extraction with multiple statistical models to achieve song categorization by genre,” in Proc. Audio Engineering Society 122nd Int. Conv., (Vienna, Austria), Audio Engineering Society, May 2007.
• B.Fields, “On the viability of using mixed feature extraction with multiple statistical models to achieve song categorization by genre,” Master’s thesis, University of Miami, Coral Gables, FL, May 2006.