|
Accession number;03A0822723
|
| Title;Context-aware Predictive Text Input Method Using Dynamic Abbreviation Expansion |
| Author;
KOMATSU HIROYUKI
(Tokyo Inst. of Technol.)
TAKABAYASHI SATORU
(National Inst. Advanced Industrial Sci. and Technol., JPN)
MASUI TOSHIYUKI
(National Inst. Advanced Industrial Sci. and Technol., JPN)
|
Journal Title;Transactions of Information Processing Society of Japan
|
Journal Code:Z0778A
|
ISSN:0387-5806
|
|
VOL.44;NO.11;PAGE.2538-2546(2003)
|
| Figure&Table&Reference;FIG.11, TBL.7, REF.10 |
| Pub. Country;Japan |
| Language;Japanese |
| Abstract;Predictive text input systems reduce the cost of input bypredicting user's input using the knowledge of natural languages and the history of user's operations. In recent years, the predictive text input systems have become popular especially for mobile phones and mobile information appliances. Since this kind of predictive input systems usually use the statistics of natural languages and the history of user's operations, there is a problem that words unused and uncommon are not likely to be chosen for the candidate words. For example, the problem becomes clear if users are doing communication using email or instant messengers because they are more likely to use unused or uncommon words according to their topics. In this paper, we propose a new predictive input method to realize a better prediction by solving the problem using context information of the peer's message. (author abst.) |
|
|
|
Related Articles;
|
