Ethernet is the most widely used network and many analytical models were developed to predict its capacity. Usually in these models assumptions like infinite population, no back-of...
A model-based clustering method is proposed for clustering individuals on the basis of measurements taken over time. Data variability is taken into account through non-linear hier...
We develop a similarity measure to detect repeatedly occurring Out-of-Vocabulary words (OOV), since these carry important information. Sub-word sequences in the recognition output...
Mirko Hannemann, Stefan Kombrink, Martin Karafi&aa...
This paper addresses the issue of optimal scale selection for circular edge extraction in the context of higher dimensional multiscale edge extraction. Based on a classification o...
: The evolution of software systems often produces incompatibilities with existing data and applications. To prevent incompatibilities, changes have to be wellplanned, and develope...