Inspiration
Mistakes are the portal to discovery
Currently, due to a large number of students in every class, it's very challenging for faculties to give individual feedback to each and every student. There are no customized study plans for students based on their performance on every test they take.
What it does
Insightium is a platform where students can get custom feedback on the tests they take. This is done through detailed feedback on the tests they take and with handouts and questionnaires on the topics where they are weak in.
How we built it
We built Insightium using Vision language models to understand the answer sheet and based on the topics where the student failed, we used language models to generate handouts and a questionnaire for further improvement in learning.
Challenges we ran into
One major challenge was ensuring user privacy and data security while accessing student's records. Another hurdle was to generate a video where an example of a topic will be explained by a kid's favorite celebrity.
Accomplishments that we're proud of
We're particularly proud of developing a user-friendly interface that is very easy to use. Additionally, the accuracy of our model to evaluate the answer sheets and the speed at which our system generates the handouts and questionnaires for the students.
What we learned
This project taught us a great deal about machine learning, image processing, and user experience design. We also gained insights into the complexities of balancing personalization with privacy.
What's next for Insightium
Looking forward, we plan to expand Insightiums's capabilities to include the parents portal where they will be able to generate vacation-based learning assignments on the fly.
Built With
- elevenlabs
- gpt4
- mistral
- openai
- python
- streamlit
- synclabs
Log in or sign up for Devpost to join the conversation.