Inspiration
Former competitions that had emerged with successful competitors who crafted outstanding solutions and the opportunities presented by AI for rapid idea prototyping.
What it does
Answers questions on Eco-Health in Nigeria using an client-side AI system that works based on keyword matches.
How we built it
We built this solution using Bolt. Then, developed the lightweight AI model and gave it a very small database of 20 datasets, which comprise tags, and descriptive text per topic to start with.
Challenges we ran into
- Problems arose when pushing the changes from our local server to the live Github Pages. Images were missing due to path errors.
- The AI's database was relatively small and we did not have enough time to generate reliable datasets.
- Due to the small dataset, some responses are a bit off point.
Accomplishments that we're proud of
- Our AI model, the Eco-Bot - typically, a static website with professional AI such will connect to a running AI service provider. By having created a client-side solution, it gave us much more opportunities to learn the concepts behind AI systems and develop it ourselves.
- The beautiful layout of the website.
What we learned
How to make our own lightweight AI model for a static page.
What's next for Eco-Health AI Assistant
Expanding the AI's databases and educating users further on the importance of Eco-Health.
Log in or sign up for Devpost to join the conversation.