SenseLearn
Education Use Case
Background
Singapore University consistently ranks among the world’s top 15 universities. During the COVID-19 pandemic, the University decided to move 80% of its curriculum online.
Objectives
Detect students’ emotional state while measuring individual student progress in learning across the curriculum.
Gather real-time insights concerning receptiveness to content and make adjustments for improvement.
Gain insights into the emotional state of students during enrolment interviews and counselling sessions.
Solutions
Our multimodal emotion recognition AI tracks students’ eye movements, facial expressions, vocal tones and vocal patterns to measure the intensity of concentration.
The system reveals when students are bored, happy, anxious, interested, angry or surprised.
This sheds light on the student’s response to the teaching content and measures reactions such as hesitation, focus and stress.
The University can use SenseLearn in individual or group settings.
Our emotion sensing algorithm is equally effective at capturing emotions in large crowd settings and in diverse conditions where background noise and changes in lighting can have an impact on concentration.
The system uses facial recognition and eye-tracking to generate data visualization of student concentration at different moments during a class.
Accurate information concerning emotional responses enables teachers to assess their own effectiveness and to improve student experience by adjusting methods, engagement, style, or content.
Benefits
Accurately assessing a student’s attention level lets a teacher know which content the students find difficult to understand.
A teacher can repeat or clarify the section or improve the content.
The emotional data can also be combined with other data such as class participation to improve on the accuracy of evaluations of student performance.
For example, a student who scored as focused 94% of the time in the class, but only manages to answer one of the teacher’s questions in a week might be considered to have low class participation.
Annotated data may shed light on issues that need to be addressed. Emotion recognition AI can provide valuable insights into the reasons for low course completion and reduced participation in courses.
As students vary in their ability to learn, emotion sensing can assist teachers in detecting, identifying and responding to a specific student’s individual needs.
Emotions like sadness and fear can help identify students experiencing learning difficulties or who are significantly falling behind.
Early intervention may help a student overcome personal difficulties.