OkenReader: ML-based classification of the reading patterns using an Apple iPad
В. Н. Анисимов, K. Chernozatonsky, Andrey V. Pikunov, M. Raykhrud, A. Revazov, K. Shedenko, D.D. Zhigulskaya, S. Zuev · Procedia Computer Science · 2021
Digital learning and professional training require processing of large volumes of information, mostly in the text format. The new information environment requires the development of new studying methods and, most importantly, a new tool for assessing its quality in order to adjust the strategy of compiling the educational content. Management and control of perception in reading can be performed with the use of technology, assessing attention, engagement, understanding, cognitive load and tiredness of readers. The article describes an early effort towards creation of a commercial technology based on a machine-learning (ML) algorithm that uses readers’ eye-tracking recording as a proxy for their cognitive state. The solution was realized on Ipad Pro with a standard iOS operating system, which provides opportunities for mass adoption in educational and training settings.