Abstract

Huge volumes of complex data time series have been generated in different domains, demanding the definition of effective and efficient services for appropriate storage and retrieval. The term complex data refers to the universe of potentially non-structured multimodal data, and therefore covers a wide variety of data types such as text, sound, image, video, graphs, geographic information, among others.  This lecture will address research problems and recent research initiatives related to the specification and implementation of appropriate systems to handle large-scale complex data time series collections, especially in the areas of feature extraction, machine learning, and data fusion. Three important applications will be addressed: Information Retrieval, Multimedia Classification, and e-Science.