Problem Statement
During live lectures, students can submit multiple comments and doubts. However, when the number of comments grows, teachers must manually read every submission to understand the common areas of confusion. This becomes difficult in large classrooms and may cause important concerns to be overlooked.
Proposed Solution
Introduce an AI-powered doubt clustering system that automatically groups similar student comments into common topics.
For example, comments such as:
- "Can you explain inheritance again?"
- "Didn't understand parent and child classes."
- "What is inheritance in OOP?"
would be grouped under a single topic:
🔴 Inheritance (23 mentions)
The analytics dashboard should display:
- Most discussed confusion topics
- Number of students associated with each topic
- Topic severity based on frequency
- Expandable view to inspect related comments
This will help teachers quickly identify major learning gaps without reading every individual comment.
Alternative Approaches
A simple keyword matching system could be used, but it may fail when students use different wording for the same concept. Topic clustering provides more accurate grouping and a better analytics experience.
Affected Area
Mockups / Additional Context
Example Dashboard:
Top Confusion Topics
🔴 Inheritance (23 students)
🟡 Polymorphism (11 students)
🟢 Constructors (4 students)
This feature would significantly improve post-lecture analysis and teacher decision-making.
Problem Statement
During live lectures, students can submit multiple comments and doubts. However, when the number of comments grows, teachers must manually read every submission to understand the common areas of confusion. This becomes difficult in large classrooms and may cause important concerns to be overlooked.
Proposed Solution
Introduce an AI-powered doubt clustering system that automatically groups similar student comments into common topics.
For example, comments such as:
would be grouped under a single topic:
🔴 Inheritance (23 mentions)
The analytics dashboard should display:
This will help teachers quickly identify major learning gaps without reading every individual comment.
Alternative Approaches
A simple keyword matching system could be used, but it may fail when students use different wording for the same concept. Topic clustering provides more accurate grouping and a better analytics experience.
Affected Area
Mockups / Additional Context
Example Dashboard:
Top Confusion Topics
🔴 Inheritance (23 students)
🟡 Polymorphism (11 students)
🟢 Constructors (4 students)
This feature would significantly improve post-lecture analysis and teacher decision-making.