ITE Typing dataset

A large-scale mobile typing dataset contains 46 755 participants typing sentences in English and 8661 participants in Finnish on their own mobile devices. Participants used various iPhone and Android devices with different operation system versions. The data was collected between 2019 and 2020 by the Computational Behaviour Lab of Aalto University. The user's typing operations and use of Intelligent Text Entry (ITE) methods (Autocorrection and Suggestion Bar) are labelled on a keystroke level. The dataset enables analysis of the effects of the user demographics and the usage and accuracy of ITE methods on typing. The dataset also has a separate table for all ITE corrected and predicted words e.g. for the ITE error analysis. Code repository: https://github.com/aalto-speech/ite-typing-dataset/ Citation: Leino, Katri, Markku Laine, Mikko Kurimo, and Antti Oulasvirta. Mobile Typing with Intelligent Text Entry: A Large-Scale Dataset and Results. 2024. https://doi.org/10.21203/rs.3.rs-4654512/v1本大规模移动键入数据集涵盖46755名使用英语键入语句的参与者，以及8661名使用芬兰语的参与者，所有参与者均使用自有移动设备完成键入任务。参与者所使用的设备涵盖多款iPhone与Android设备，且系统版本各不相同。本数据集于2019年至2020年间由阿尔托大学（Aalto University）计算行为实验室采集完成。 用户的键入操作与智能文本输入（Intelligent Text Entry，简称ITE）方法（包括自动校正（Autocorrection）与建议栏（Suggestion Bar））的使用情况均已在按键层级完成标注。本数据集可用于分析用户人口统计学特征、智能文本输入方法的使用情况与准确率对键入行为的影响。此外，数据集还设有独立表格，收录所有经ITE校正与预测的词汇，可用于智能文本输入的错误分析等研究场景。 代码仓库：https://github.com/aalto-speech/ite-typing-dataset/ 引用： Leino, Katri、Markku Laine、Mikko Kurimo与Antti Oulasvirta. 《基于智能文本输入的移动键入：大规模数据集与研究成果》，2024年。https://doi.org/10.21203/rs.3.rs-4654512/v1