Scalable Machine Learning for Fast Thematic Mapping in Web Servers

@Article {
author = {Javier Lozano, Naiara Aginako, Marco Quartulli, Ekaitz Zulueta, Igor García Olaizola},
title = {Scalable Machine Learning for Fast Thematic Mapping in Web Servers},
pages = {38-41},
keywds = {

data processing, data collection, information storage and retrieval, information processing, information technology, information technology applications, document retrieval, computer system, space research, European Space Agency, European Union Satelli
}
abstract = {

We present a web mining processing service that returns supervised probabilistic classifications of Earth Observation (EO) data in tiled form, with the aim to create user-selection based thematic maps from remotely sensed raster imagery. User interfaces supporting interactive navigation and model training and tuning are implemented in open HTML5 standards, while software interfaces among components conform to OGC standards. Near real time operation in the servers is attained by exploiting efficient data structures for high dimensional indexing and search.

}
isbn = {978-92-79-43252-1},
date = {2014-11-12},
year = {2014},
}