Please use this identifier to cite or link to this item:
Title: 替代螢幕鍵盤與常用字預測功能對重度肢體障礙者中文輸入成效之研究
Designing and Evaluating Chinese Scanning Technique on an Alternative On-screen Keyboard with Word Prediction for Individuals with Severely Physical Impairment
Authors: 李天佑
Keywords: 螢幕鍵盤
on-screen keyboard
word completion
word prediction
Issue Date: 2006
Abstract: 本研究的目的在於設計出替代螢幕鍵盤與常用字預測功能並且探討這兩項介入對重度肢體障礙者在中文輸入上的成效。研究者選取八十七年常用語詞調查報告中的640個字,將這640個字各自拆解出嘸蝦米字根並分析每個嘸蝦米字根的出現率,根據出現率重新排列標準螢幕鍵盤版面,使其成為6列*7行的替代螢幕鍵盤。 接下來,研究者同樣利用這640個字當作常用字預測資料庫,實作出常用字預測功能。讓使用者可以在輸入每個中文字的前兩個嘸蝦米字根後,自動產生常用字列表供使用者選取。 替代螢幕鍵盤與常用字預測功能設計完成之後,研究者利用實驗來探討此兩項介入對重度肢體障礙者在中文輸入上的成效。實驗採單一受試者交替處理,實驗對象為二位脊髓損傷患者與一位手部變形,喪失手指的重度肢體障礙者。實驗結果顯示,替代螢幕鍵盤與加入常用字預測的標準螢幕鍵盤相對於標準螢幕鍵盤的確有較高的中文輸入成效。
The purposes of this study were to design Chinese Scanning Technique on an alternative on-screen keyboard with word prediciton and to evaluate their effectiveness. Researcher disassembled the 640 words that were 80% of the 5063 common Chinese words into Wu-Shia-Mi roots, and then reorganized the layout of the standard keyboard into a 6 rows * 7 columns frequency-optimized alternative eyboard. When designing word prediction, the researcher used the same 640 words as database for predicting. Word prediction provides the user of on-screen keyboard with predictive suggestions after the user has entered the initial two letters of a word. In this research, there were 3 subjects with Severely Physical Impairment, participated the usability evaluation of the alternative on-screen keyboard and word prediction. The performance of the 3 subjects showed that the alternative on-screen keyboard and standard on-screen keyboard with word prediction were better than standard on-screen keyboard.
Other Identifiers: G0069108004
Appears in Collections:學位論文

Files in This Item:
File Description SizeFormat 
800401.pdf152.55 kBAdobe PDFView/Open
800402.pdf212.91 kBAdobe PDFView/Open
800403.pdf166.36 kBAdobe PDFView/Open
800404.pdf512.53 kBAdobe PDFView/Open
800405.pdf114.19 kBAdobe PDFView/Open
800406.pdf199.96 kBAdobe PDFView/Open

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.