Hybrid brain computer interface for relaxation management
Learning performance is an important feature that needs to be evaluated in managing learning performance of the students. In the past, teachers evaluate students by monitoring their grades. However, now we are more concerned with not only just grades but also the engagement level of students. A BCI...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Final Year Project |
| 語言: | English |
| 出版: |
2017
|
| 主題: | |
| 在線閱讀: | http://hdl.handle.net/10356/71583 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Learning performance is an important feature that needs to be evaluated in managing learning performance of the students. In the past, teachers evaluate students by monitoring their grades. However, now we are more concerned with not only just grades but also the engagement level of students. A BCI with various sensors is designed to obtain information on affective state and attention level of the participants. The participants, consisting of 7 male and 3 female, are asked to complete classifier training phase first. Next, in E-learning stimulation phase, they have to complete liked and disliked passage respectively. During the experiment, electroencephalogram (EEG), eye positions, photoplethysmogram (PPG), galvanic skin response (GSR) and heart rate variability (HRV) of the participants are recorded. Data analysis will be conducted to extract out each individual feature. Classification of above 90% is achieved. Results have revealed that mental engagement is similar while visual engagement varies across different participants during the two comprehension passages. In mental engagement and mental response, participants are observed to be having lower engagement level and higher stress level during least liked passage and vice versa for most liked passage. This proved that multimodal system makes evaluation process much more reliable together with visual and mental engagement and other modalities such as heart rate variability, mental and reaction response. To conclude, the different modalities of sensors showed to have contributed to a more significant student engagement index. This would help the educators in better managing in student learning performance, enabling more meaningful improvement in learning materials. |
|---|