Home // SENSORCOMM 2017, The Eleventh International Conference on Sensor Technologies and Applications // View article

Discrimination of Immersion in Writing from Association of Browsed Information

Authors:
Takahisa Oe
Shinya Yonekura
Hiromitsu Shimakawa

Keywords: Sensorless detection; Immersion; Distraction; Machine learning

Abstract:
This study proposes a method to distinguish student immersion in writing on computers without physical sensors. To improve the work efficiency of writing, students need support, such as warning when they have been distracted a long time from writing. A model using the Random Forest algorithm discriminates the immersion, examining windows of their operation target on the top of the display. In our experiment, the model discriminates the immersion of 5 subjects with the accuracy of 0.65 or higher in the F-measure, where placement of a specific window on the top of the screen turns out to be the most important feature. Various kinds of information is presented on the screen of the PCs of students. It includes not only information necessary for writing, but also entertainment information such as movies and games. The experiment result indicates that students tend to exclude entertainment information from their vision when they are under immersion in writing. It suggests that the student distraction from writing can be warned without extra effort from students, if we examine the top of the screen.

Pages: 64 to 70

Copyright: Copyright (c) IARIA, 2017

Publication date: September 10, 2017

Published in: conference

ISSN: 2308-4405

ISBN: 978-1-61208-580-7

Location: Rome, Italy

Dates: from September 10, 2017 to September 14, 2017