| Download | - View final version: Visualizing learning performance data and model predictions as objects in a 3D space (PDF, 535 KiB)
|
|---|
| Author | Search for: Emond, Bruno1; Search for: Valdés, Julio J.1 |
|---|
| Affiliation | - National Research Council of Canada. Digital Technologies
|
|---|
| Format | Text, Article |
|---|
| Conference | Proceedings of the 12th International Conference on Educational Data Mining, July 2-5, 2019, Montréal, Quebec |
|---|
| Subject | visual analytics; dimensionality reduction; closing the loop between research and practice |
|---|
| Abstract | This poster/demonstration presents our preliminary efforts to explore learning data and model predictions by mapping them into objects in a three-dimensional space. Our work is related to the area of visual analytics, where machine learning or other analytic techniques are combined with interactive data visualization to promote sense making and analytical reasoning. The demonstration consists of interactive VRML models generated from the DataShop Geometry9697 dataset. The initial results indicate that in spite of some limitations the approach allowed to identify from only the observations distribution specific knowledge components that could be targeted for model refinement. |
|---|
| Publication date | 2019-07-05 |
|---|
| Publisher | International Educational Data Mining Society |
|---|
| In | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | a27cef6d-0502-48dd-8625-855cb21c6440 |
|---|
| Record created | 2021-07-16 |
|---|
| Record modified | 2021-07-16 |
|---|
