Soft Keyboard Performance Optimization

Xiaojun Bi, Brian A. Smith, Tom Ouyang, Shumin Zhai · Oxford University Press eBooks · 2018

Optimization techniques have played a vital role in improving the performance (i.e., input speed and accuracy) of soft keyboards. This chapter introduces the challenges, methodologies, and results of keyboard performance optimization. Leveraging the robust human motor control phenomena manifested in text entry, we used the Metropolis random walk algorithm, and Pareto multi-objective optimization method to optimize the keyboard layout and a soft keyboard decoder. The optimization led to layouts that shorten finger travel distance and improve the input speed as well as accuracy over the Qwerty layout, and a soft keyboard decoder with improved correction and completion ability.

Read the paper · More papers on PaperTik