The signal acquisition was carried out using a smartwatch and included real biosignal data such as acceleration, heart rate, and other metadata. In addition, emotions and context were recorded every 15 minutes. In total, data from 48 + 5 participants were collected at four different job fairs. The project was developed in collaboration with Professor Mädche’s research group at the Karlsruhe Institute of Technology (Human-Centered Systems Lab).
































The task consists of predicting emotions and context based on the recorded biosignals. In the Student Track it is a person-specific task in which predictions are made for individuals for whom at least two known labels are available. In the Professional Track on the other hand, predictions should also be made for individuals for whom no labels are available.
| Rank | Score | Team Name | Member Name(s) |
|---|---|---|---|
| 🥇1 | 0.585 | Shallow Learning | Ilkan Kandaz, Bjarne Kahrs |
| 🥇1 | 0.585 | Mastschweine | Felix Drees, Tjado Edzards, Leon Frölje |
| 🥉3 | 0.575 | Marshal | Arne Winter |
| 4 | 0.569 | Data Alchemist | Tobias Fricke |
| 5 | 0.549 | liuge | Henry Wangte |
| Rank | Score | Team Name | Member Name(s) |
|---|---|---|---|
| 🥇1 | 0.551 | Watson | Sören Sparmann, Lutz Terfloth |
| 🥈2 | 0.532 | Slightly Shallower Learning | Ilkan Kandaz, Bjarne Kahrs |
| 🥈2 | 0.532 | Spider Bobs | Johannes Falk, Eda Cakir, Ali Salehzadeh-Yazdi |
| 4 | 0.517 | JAAIants | Tim Cofala, Marcel Plutat, Bernd Poppinga |
| 5 | 0.513 | KTLM | Louis Mozart Kamdem, Leonie Nora Siege |
| Rank | Score | Team Name | Member Name(s) |
|---|---|---|---|
| 🥇1 | 0.647 | el jorno | Jorma Reiners |
| 🥈2 | 0.373 | Spider Bobs | Johannes Falk, Eda Cakir, Ali Salehzadeh-Yazdi |