Word space models, in the sense of vector space models built on distributional data taken from texts, are used to model semantic relations between words. We argue that the high dim...
Recognition and retrieval of historical handwritten material is an unsolved problem. We propose a novel approach to recognizing and retrieving handwritten manuscripts, based upon ...
We introduce new techniques for extracting, analyzing, and visualizing textual contents from instructional videos of low production quality. Using Automatic Speech Recognition, ap...
Semistatic word-based byte-oriented compression codes are known to be attractive alternatives to compress natural language texts. With compression ratios around 30%, they allow di...
Words of foreign origin are referred to as borrowed words or loanwords. A loanword is usually imported to Chinese by phonetic transliteration if a translation is not easily availa...