In the context of an experimental virtual-reality surgical planning software platform, we propose a fully self-assessed adaptive region growing segmentation algorithm. Our method s...
In China-US Million Book Digital Library, output of the digitalization process is more than one terabyte of text in OEB and PDF format. To access these data quickly and accurately,...
We propose a novel approach to crosslingual language model (LM) adaptation based on bilingual Latent Semantic Analysis (bLSA). A bLSA model is introduced which enables latent topi...
Words in Semitic texts often consist of a concatenation of word segments, each corresponding to a Part-of-Speech (POS) category. Semitic words may be ambiguous with regard to thei...
In order to solve problems of reliability of systems based on lexical repetition and problems of adaptability of languagedependent systems, we present a context-based topic segmen...