IEICE Transactions on Information and Systems
Online ISSN : 1745-1361
Print ISSN : 0916-8532
Regular Section
Predicting Uninterruptible Durations of Office Workers by Using Probabilistic Work Continuance Model
ジャーナル フリー

2020 年 E103.D 巻 4 号 p. 838-849


In order not to disrupt a team member concentrating on his/her own task, the interrupter needs to wait for a proper time. In this research, we examined the feasibility of predicting prospective interruptible times of office workers who use PCs. An analysis of actual working data collected from 13 participants revealed the relationship between uninterruptible durations and four features, i.e. type of application software, rate of PC operation activity, activity ratio between keystrokes and mouse clicks, and switching frequency of application software. On the basis of these results, we developed a probabilistic work continuance model whose probability changes according to the four features. The leave-one-out cross-validation indicated positive correlations between the actual and the predicted durations. The medians of the actual and the predicted durations were 539 s and 519 s. The main contribution of this study is the demonstration of the feasibility to predict uninterruptible durations in an actual working scenario.

© 2020 The Institute of Electronics, Information and Communication Engineers
前の記事 次の記事