Disaster Risk Analytics Platform

Location - A8

Our team of has developed a risk analysis platform designed specifically for investment firms operating in the U.S. This tool leverages advanced statistical analysis and deep learning technologies to predict potential disasters and their financial impacts in various U.S. states, providing firms and also customers with the insights needed to make informed, resilient investment decisions. Our platform offers a strategic advantage by transforming uncertainty into a calculable risk, enhancing the ability to plan and prosper in volatile markets.

Our risk analysis platform harnesses the power of advanced statistical and machine learning techniques to provide comprehensive disaster risk assessments for investment firms focusing on U.S. markets. At the core of our analytics, we utilize Poisson's distribution to model the frequency of disaster events, offering a solid statistical foundation for understanding risk probabilities. To delve deeper into predicting potential financial impacts, we've integrated cutting-edge deep learning technologies. Our platform features an autoencoder for efficient data encoding and dimensionality reduction, ensuring robust feature extraction. We have also developed a custom deep learning based regression model, named "Regression Net," which specializes in financial damage estimation based on the processed data. The user interface of our platform is built as a Flask web application, showcasing an intuitive map interface that allows users to interact with and visualize the data seamlessly. This app has been deployed on AWS EC2, ensuring scalable and reliable access. The integration of these technologies makes our platform not only innovative but also extremely practical and user-friendly for investment firms looking to mitigate financial risks associated with natural disasters.

Flask, HTML, CSS,JavaScript, Machine Learning, Deep Learning, Autoencoders, RegressionNet


Subhrato Som (ss5654@drexel.edu) ; Imon Bera (ib385@drexel.edu) ; Abhishek Baj (ab4777@drexel.edu)

Selected Prizes

  • Making insurance services equally available and accessible to all. Prize Apple Air Pods

  • Use Open AI create Chat Bot assistant to support customers and employees with Comcast products. Winners will receive a trip to the Comcast Innovation Center along with lunch, tour and meeting with Comcast executives

  • Leveraging any Comcast Product or Service use AI to support for the common good the aging at home population. Winners will receive a trip to the Comcast Innovation Center along with lunch, tour and meeting with Comcast executives

  • Your project must represent principles of diversity, equity, inclusion, and/or belonging