Paper Abstract and Keywords |
Presentation |
2010-03-12 11:00
Validity analysis of statistical learning -based Japanese word segmentation for braille translation Aki Sugano, Mika Ohta, Kenji Miura (Kobe Univ.), Masako Matsuura (Kobe Univ. Hosp.), Yuji Matsumoto (NAIST), Toshiko Ohshima (Kobe Univ. Hosp.), Yutaka Takaoka (Kobe Univ.) WIT2009-82 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
To provide medical information in Braille to the visually impaired patients, we are developing Japanese-into- Braille translating program eBraille for medicine (eBraille-M). Its translation engine named KUIC includes braille transcription rules based on Japanese part of speech. The braille transcription rules other than that for part of speech are word segmentation by using word sense and pronunciations. To include such rules for our program, we developed a statistical learning model to analyze its effectiveness for braille translation. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
Braille / medical terms / statistical learning model / / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 109, no. 467, WIT2009-82, pp. 5-8, March 2010. |
Paper # |
WIT2009-82 |
Date of Issue |
2010-03-05 (WIT) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
WIT2009-82 |
|