A new approach to the recognition of temporal behaviors and activities is presented. The fundamental idea, inspired by work in speech recognition, is to divide the inference probl...
Background: Channel current feature extraction methods, using Hidden Markov Models (HMMs) have been designed for tracking individual-molecule conformational changes. This informat...
Stephen Winters-Hilt, Matthew Landry, Mark Akeson,...
A standard form of analysis for linguistic typology is the universal implication. These implications state facts about the range of extant languages, such as “if objects come af...
Abstract. We present an exploratory method for simultaneous parcellation of multisubject fMRI data into functionally coherent areas. The method is based on a solely functional repr...
We present a novel paradigm for statistical machine translation (SMT), based on a joint modeling of word alignment and the topical aspects underlying bilingual document-pairs, via...