TUW @ Retrieving Diverse Social Images Task 2014

J. Palotti, N. Rekabsaz, M. Lupu, A. Hanbury:
"TUW @ Retrieving Diverse Social Images Task 2014";
Vortrag: MediaEval Benchmarking Initiative for Multimedia Evaluation, Barcelona, ES; 16.10.2014 - 17.10.2014; in:"Working Notes Proceedings of the MediaEval 2014 Workshop", CEUR-WS, Vol-1263 (2014), ISSN: 1613-0073.

[ Publication Database ]

Abstract:


This paper describes the e orts of Vienna University of Technology (TUW) in the MediaEval 2014 Retrieving Diverse Social Images challenge. Our approach consisted of 3 steps: (1)
a pre-filtering based on Machine Learning, (2) a re-ranking
based on Word2Vec, and (3) a clustering part based on an
ensemble of clusters. Our best run reached a F@20 of 0.564.