|
|
Xavier
Sevillano Domínguez |
||
|
|
|
|
|
|
Before enumerating the publications that have originated from my research activities, a few words about how it has evolved from 2001 to present. My first research steps were oriented towards the field of speech enhacement. In particular, I worked on the use of adaptive filtering techniques for noise reduction, an application known as adaptive noise cancellation. Moreover, I also studied the use of blind source separation techniques, such as Independent Component Analysis (ICA), for separating and locating speakers in multi-microphone environments. In 2003, the direction of my research activities suffered a notable shift, as I started working on text analysis, focusing on its application on multi-domain text-to-speech synthesis systems. In particular, I worked on the design of hierarchical text classifiers, using ICA for feature extraction. Given the unsupervised nature of ICA, I focused on the problem of unsupervised text classification (or text clustering), analyzing the behaviour of multiple feature extraction techniques (such as Principal Component Analysis, Non-negative Matrix Factorization or Random Projection) in this context. Given the lack of inherent superiority of none of these feature extraction techniques, I started to think of ways of combining the outcomes of multiple unsupervised classifiers, a task called consensus clustering, also known as cluster ensembles. In parallel, I also studied the application of feature extraction techniques for multimedia data processing (analysis and synthesis), putting special emphasis on the problem of multimedia data clustering. From June to December 2011 I was a visiting researcher at the Multimedia & Vision Research Group of the Queen Mary, University of London. My work there was focused on multimedia indexing, with application to geo-tagging video resources. Click on the following link in order to see a list of my publications sorted by subject. Aside from that, I have been a program committee member and/or reviewer for the following international conferences and journals: ·
International
Conference on Industrial, Engineering & Other Applications of Applied
Intelligent Systems (IEA) 2010 (Special
session on Ensemble Learning) · IEEE International Conference on Multimedia and Expo (ICME) 2011, 2012 · IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) 2012 · European Conference on Computer Vision (ECCV) 2012 (Workshop on Information Fusion for Concept Recognition) · IEEE International Conference on Machine Learning and Applications (ICMLA) 2012 (Special session on Machine Learning with Multimedia Data) · IEEE Transactions on Multimedia · IEEE Transactions on Neural Networks and Learning Systems · Elsevier Pattern Recognition Letters · Elsevier Knowledge Based Systems
|
|||