Reddingius, Peter (2020) Combining online and offline machine learning to create an interruption management system using eye data. Master's Thesis / Essay, Human-Machine Communication.
|
Text
Master's thesis Peter Reddingius (s3272222).pdf Download (1MB) | Preview |
|
![[img]](https://fse.studenttheses.ub.rug.nl/style/images/fileicons/text.png) |
Text
toestemming.pdf Restricted to Registered users only Download (116kB) |
Abstract
In our current COVID-19 society everyone is advised to work from home as much as possible. That can lead to many new problems, such as a whole new set of possible distractions and interruptions. Interruptions have been shown to be disruptive, which can lead to errors and stress. Interruptions during low workload moments are much less disruptive than during high workload moments. Unfortunately, interruptions do not usually take the current workload into account. Using pupil dilation, a measure of working memory usage which is strongly related to workload, better moments for interruptions can be found. This paper proposes a task-independent machine learning approach to manage interruptions at low workload moments. A mondrian forest classifier was pre-trained on eye data from participants of another interruption experiment, and it also continues learning while the participant performs the experiment. To test the viability of the interruption management system (IMS) the paper also proposes a new experimental task to test the viability of an IMS using the game of sudoku. The proposed experimental task uses sudokus which provide a clear distinction between high and low workload moments. While the experiment could not be completed due to the pandemic, a small pilot experiment showed promising results. The IMS seems to be able to predict moments of low workload and managed to improve at finding good moments for interruptions while the participant was performing the experiment.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Supervisor name: | Borst, J.P. |
| Degree programme: | Human-Machine Communication |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 12 Jan 2021 10:00 |
| Last Modified: | 12 Jan 2021 10:00 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/23794 |
Actions (login required)
 |
View Item |