Text and context in mobile computing

移动计算中的文本和上下文

• 批准号: 121581-2007
• 负责人: MacKenzie, Scott
• 金额: $ 1.42万
• 依托单位: York University
• 依托单位国家: 加拿大
• 项目类别: Discovery Grants Program - Individual
• 财政年份: 2007
• 资助国家: 加拿大
• 起止时间: 2007-01-01 至 2008-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=361151
• 关键词: Text; context; mobile; computing

A program of research is proposed that will focus on "text and context in mobile computing". By "text", we refer simply to the issue of text entry. With over 1 billion SMS messages sent per day worldwide, the mobile phone keypad is now one of the dominant devices for text entry - this despite its poor capabilities for such. And so, a program of research in mobile text entry is extremely worthwhile. By "context", we refer to the many possible mechanisms to aide text entry through embedded knowledge of the language, user, or device. Word prediction, word completion, suffix completion, are some examples, but the general problem of interfacing with ambiguous keyboards (e.g., the mobile phone keyboard has either 3 or 4 letters per key) is of particular interest. As mobile devices shrink in size, there is increasing interest in supporting text entry with fewer keys than even a mobile phone contains. This presents an interesting combination of problems, intermixing issues of ambiguity resolution with an embedded language model. There is a tension between increasing keystrokes for ambiguity resolution (as the number of keys decreases) and decreasing keystrokes using predictive aides. This is fertile ground for research and it is the primary focus of the program of research proposed. The PI also works extensively with computer pointing devices, models of interaction (e.g., using Fitts' law), and the measurement of human performance. These topics are also part of the proposed program of research.

提出了一个研究计划，将集中在“文本和上下文在移动的计算”。我们所说的“文本”，只是指文本输入的问题。由于全世界每天发送超过10亿条SMS消息，移动的电话键盘现在是用于文本输入的主要设备之一-尽管其用于文本输入的能力很差。因此，在移动的文本输入的研究计划是非常值得的。通过“上下文”，我们指的是通过语言、用户或设备的嵌入知识来辅助文本输入的许多可能机制。单词预测、单词补全、后缀补全是一些示例，但是与模糊键盘（例如，移动的电话键盘每个键具有3或4个字母）是特别感兴趣的。随着移动的设备尺寸的缩小，人们越来越关注用比移动的电话包含的键还少的键来支持文本输入。这提出了一个有趣的问题组合，将歧义解决问题与嵌入式语言模型混合在一起。在增加用于模糊度解决的模糊度系数（随着键的数量减少）和使用预测辅助来减少模糊度系数之间存在紧张关系。这是研究的沃土，也是研究计划的主要重点。PI还广泛地与计算机定点设备、交互模型（例如，使用Fitts定律），以及人类表现的测量。这些主题也是拟议研究计划的一部分。