Data abstraction and retrieval in Text and Image documents
Date: 28/10/09
Venue: MS020
Time: 4.00 pm
Speaker: Dr. Mourad Oussalah
Affiliation: University of Birmingham, UK
Data abstraction and retrieval in Text and Image documents
By
Dr. Mourad Oussalah
University of Birmingham, UK
Abstract
As the size of electronic documents tend to increase drastically with a variety of data type including text, image, song, building efficient information retrieval systems becomes a crucial task in information processing community. The talk describes some of our work in text summarization both extracted and abstracted summarization task using Wikepedia and WordNet corpus. Especially approaches dealing with semantic aspects of the text will be detailed. Next an analogy is built with respect to image retrieval, where the face recognition task using Haar features, PCA (Principal Component Analysis) and NMF (Negative Matrix Factorization) will be presented.
Short Biography
Dr. Mourad Oussalah holds a PhD degree in robotics and artificial intelligence from University of Paris XII in France in 1998. He then worked at KU Leuven in Belgium in Active Sensing project as a Research Fellow, then in 2000 he moved to City University of London where he worked in DIRC multidisciplinary project on dependability of human-machine organizations. Since 2003, he is lecturing at university of Birmingham in Electronics, Electrical and Computer Engineering. His topics of interest include data mining and information retrieval, location based systems, target tracking and computer vision
