Klaas Bosteels

Email: firstname.lastname@gmail.com (substitute where necessary)

Short biography

Klaas Bosteels obtained a M.Sc. degree in Computer Science at Ghent University, Belgium, with a thesis that received the Baudouin Elleboudt Award 2006 for the best Belgian graduate thesis in the ICT domain. In October 2006, he joined the Department of Applied Mathematics and Computer Science, Ghent University, where he worked towards a Ph.D. degree as a member of the Fuzziness and Uncertainty Modelling Research Group, in close cooperation with Last.fm. He obtained his Ph.D. in December 2009, after which he joined Last.fm full time as Data and Scalability Engineer. In January 2011 he moved back to Belgium to join Massive Media as Data Scientist.

Publications

Papers in international journals listed in SCI:

A fuzzy framework for defining dynamic playlist generation heuristics
K. Bosteels, E. E. Kerre
Fuzzy Sets and Systems 160, p. 3342-3358, 2009
On a reflexivity-preserving family of cardinality-based fuzzy comparison measures
K. Bosteels, E. E. Kerre
Information Sciences 179, p. 2342-2352, 2009
» PDF, cheat sheet
A triparametric family of cardinality-based fuzzy similarity measures
K. Bosteels, E. E. Kerre
Fuzzy Sets and Systems 158, p. 2466-2479, 2007
» PDF, cheat sheet

Papers in other international journals:

A biparametric family of cardinality-based fuzzy similarity measures
K. Bosteels, E. E. Kerre
New Mathematics and Natural Computation 3, p. 307-319, 2007
» PDF

Chapters in books:

Fuzzy audio similarity measures based on spectrum histograms and fluctuation patterns
K. Bosteels, E. E. Kerre
Computational Intelligence in Multimedia Processing (A. E. Hassanien, J. Kacprzyk, A. Abraham, eds.), Studies in Computational Intelligence, p. 213-231, 2008
» PDF

Papers in proceedings:

Evaluating and analysing dynamic playlist generation heuristics using radio logs and fuzzy set theory
K. Bosteels, E. Pampalk, E. E. Kerre
Proceedings of the ISMIR International Conference on Music Information Retrieval (ISMIR'09), 2009
» PDF
On the benefits of representing music objects as fuzzy sets
K. Bosteels, E. Pampalk, E. E. Kerre
Proceedings of the 2009 IFSA-EUSFLAT Conference (IFSA-EUSFLAT'09), 2009
» PDF
Fuzzy audio similarity measures based on spectrum histograms and fluctuation patterns
K. Bosteels, E. E. Kerre
Proceedings of the 2007 International Conference on Multimedia and Ubiquitous Engineering (MUE'07), 2007
» PDF

Other publications:

Fuzzy techniques in the usage and construction of comparison measures for music objects
K. Bosteels, PhD thesis, Ghent University, 2009
Supervisor: Prof. Dr. E. E. Kerre
» PDF
Using fuzzy set theory to construct comparison measures for music objects
K. Bosteels, E. Pampalk, E. E. Kerre
Presented during the doctoral symposium of the Faculty of Scieces at Ghent University, 2009
» abstract
Music retrieval based on social tags: A case study
K. Bosteels, E. Pampalk, E. E. Kerre
Presented during the late-breaking/demo session of the ISMIR International Conference on Music Information Retrieval, 2008
» abstract, poster
Combining audio similarity measures using generalizations of logical connectives
K. Bosteels, E. E. Kerre
Implementation submitted to the 4th annual Music Information Retrieval Evaluation eXchange (MIREX), 2007
» PDF, poster
Similariteitsgebaseerd rangschikken van beelden in zoekmachines
K. Bosteels, Master's thesis, Ghent University, 2006
Supervisor: Prof. Dr. E. E. Kerre
» PDF, extended abstract, presentation, prototype

Projects

Last.fm's playground
Testing of new ideas and experimental technologies that leverage Last.fm data.
» link
Dumbo
Python module that allows you to easily write and run Hadoop programs.
» wiki, blog

Awards

Baudouin Elleboudt Award 2006
Award for the best Belgian graduate thesis in the ICT domain, granted by the Belgian association of journalists in information technology (B-JIT).