TAILIEUCHUNG - Báo cáo khoa học: "Evaluating Word Prediction: Framing Keystroke Savings"

Researchers typically evaluate word prediction using keystroke savings, however, this measure is not straightforward. We present several complications in computing keystroke savings which may affect interpretation and comparison of results. We address this problem by developing two gold standards as a frame for interpretation. These gold standards measure the maximum keystroke savings under two different approximations of an ideal language model. The gold standards additionally narrow the scope of deficiencies in a word prediction system. . | Evaluating Word Prediction Framing Keystroke Savings Keith Trnka and Kathleen F. McCoy University of Delaware Newark DE 19716 trnka@ Abstract Researchers typically evaluate word prediction using keystroke savings however this measure is not straightforward. We present several complications in computing keystroke savings which may affect interpretation and comparison of results. We address this problem by developing two gold standards as a frame for interpretation. These gold standards measure the maximum keystroke savings under two different approximations of an ideal language model. The gold standards additionally narrow the scope of deficiencies in a word prediction system. 1 Introduction Word prediction is an application of language modeling to speeding up text entry especially to entering utterances to be spoken by an Augmentative and Alternative Communication AAC device. AAC devices seek to address the dual problem of speech and motor impairment by attempting to optimize text input. Even still communication rates with AAC devices are often below 10 words per minute Newell et al. 1998 compared to the common 130-200 words per minute speech rate of speaking people. Word prediction addresses these issues by reducing the number of keystrokes required to produce a message which has been shown to improve communication rate Trnka et al. 2007 . The reduction in keystrokes also translates into a lower degree of fatigue from typing all day Carlberger et al. 1997 . Word prediction systems present multiple completions of the current word to the user. Systems generate a list of W predictions on the basis of the word being typed and a language model. The vocabulary is filtered to match the prefix of the current word and the language model ranks the words according to their likelihood. In the case that no letters of the current word have been entered the language model is the sole factor in generating predictions. Systems often use a touchscreen or function number

• Báo cáo khoa học
• Evaluating Word Prediction
• Framing Keystroke Savings
• Keith Trnka and Kathleen F
• McCoy
• báo cáo khoa học
• báo cáo ngôn ngữ
• xử lý ngôn ngữ tự nhiên
• Bài giảng Xử lý ngôn ngữ tự nhiên
• Ngôn ngữ tự nhiên
• Kỹ thuật lập trình
• Xử lý nhập nhằng ngữ nghĩa
• Xử lý ngữ nghĩa
• Xử lý ngôn ngữ
• Natural Language Processing
• Ngôn ngữ lập trình
• Nhập nhằng cấu trúc
• Tri thức về ngôn ngữ
• Mô hình ngôn ngữ
• Mô hình n gram
• Dịch máy
• Phương pháp dịch máy
• Hiểu ngôn ngữ
• Phân tích ngữ nghĩa
• Phân loại tin tự động
• Biểu diễn vị từ
• Thuộc tính về sự kiện
• Thuộc tính về sự kiện
• Mô hình ngôn ngữ Google Book N grams
• Mô hình ngôn ngữ KenLM
• Phương pháp làm mịn
• Tách từ tiếng Việt
• Gán nhãn từ loại
• Phân tích cú pháp
• Nghĩa từ vựng
• Phân giải nhập nhằng từ
• Phân loại văn bản
• Lexical chain
• Phản hồi thông tin
• Bài toán PTCP
• Cấu trúc ngữ pháp
• Phân tích cú pháp xác suất
• CKY kết hợp xác suất
• Văn phạm phi ngữ cảnh xác suất
• Hình thái học