Inspiration
A major flood hit my city in India, and I witnessed firsthand how unprepared people were. The disaster struck suddenly, with little to no warning, leaving many vulnerable and without timely help. That experience motivated us to build a system that could provide early alerts and prevent such chaos. We strongly believe that “Prevention is better than cure,” and this project is our step toward making communities safer and more prepared.
What it does
RESQnet is a disaster prediction and response system that analyzes environmental and demographic data to assess flood risk in different regions. It classifies areas into risk levels, estimates the potential impact on the population, and provides actionable recommendations such as rescue team allocation and essential supplies. The platform also includes an early warning system that alerts users about potential disasters before they occur.
How we built it
We built RESQnet using Python and Streamlit for a simple and interactive web interface. We used Pandas to handle and process data, and implemented a rule-based AI model to calculate risk scores based on factors like rainfall, population density, and river levels. The system integrates all components into a dashboard that displays predictions, alerts, and recommendations in real time.
Challenges we ran into
One of the biggest challenges was integrating map-based data and ensuring it aligned correctly with our system. Designing a clean and intuitive UI was also difficult, as we wanted it to be both functional and visually appealing. Additionally, collecting and structuring meaningful data, as well as building a reliable AI logic within a short hackathon timeframe, required significant effort and iteration.
Accomplishments that we're proud of
We are proud to have built a fully working, end-to-end disaster prediction system during our first hackathon. From idea to implementation, we successfully created a solution that is not only functional but also impactful and demo-ready.
What we learned
Through this project, we gained hands-on experience with APIs, AI logic, and web application development. We learned how to build and structure a project using VS Code and Streamlit, how to process and analyze data effectively, and how different components of an intelligent system come together in a real-world application.
What's next for RESQnet
We plan to expand RESQnet by adding individual user profiles with emergency contacts, so that alerts can be automatically sent to family members or authorities during disasters. We also aim to integrate real-time data sources, support multiple types of natural disasters, and improve the accuracy of predictions using more advanced models.
Log in or sign up for Devpost to join the conversation.