Other formats:
BibTeX
LaTeX
RIS
@misc{837859,
author = {Sojka, Petr},
address = {Brno},
edition = {první},
keywords = {competing patterns;context-sensitive patterns;machine learning;natural language engineering;hyphenation; [Thai text] segmentation;disambiguation;part of speech tagging;digitization;digital libraries;navigation;hypertext;MSC;text classification},
language = {eng},
location = {Brno},
publisher = {Masarykova univerzita},
title = {From Minds to Pixels and Back},
year = {2008}
}
TY - GEN
ID - 837859
AU - Sojka, Petr
PY - 2008
TI - From Minds to Pixels and Back
VL - habilitation thesis
PB - Masarykova univerzita
CY - Brno
KW - competing patterns;context-sensitive patterns;machine learning;natural language engineering;hyphenation
KW - [Thai text] segmentation;disambiguation;part of speech tagging;digitization;digital libraries;navigation;hypertext;MSC;text classification
N2 - This cumulative work consists of my papers in the areas of electronic document preparation---from authors' minds to the pixels on paper or screen---and digitization---from the scanned papers back to the minds of scholars via the structured digital library. In the first part of this thesis, a short overview of the papers and topics tackled is given. The second part consist of a series of eight published papers that describe problems addressed: applications of competing patterns and related learning methods in areas of hyphenation, the hyphenation of compound words and, for example, the segmentation of Thai texts. A method of single-source publishing for various output media is developed and shown. The third part consist of a series of six papers and drafts with topics in the areas of digitization, optical character recognition, digital library preparation and the automated classification of digitized papers.
ER -
SOJKA, Petr. \textit{From Minds to Pixels and Back}. první. Brno: Masarykova univerzita. xvi+209. habilitation thesis. 2008.
|
