Inspiration

The most pressing issue we face in the midst of the current pandemic is the lack of specific, personalized information, and the fear that uncertainty that it creates. Our project addresses this head-on, with bespoke recommendations and specific, but easy-to-understand visualization, all presented in a neat package.

What it does

The application aims to give you a general idea about the specific risk levels of various areas in your locality, to assess the relative risk of traveling to or through a specified location and recommend the best route. It gathers data from live information on government websites, live news updates and live footage (from CCTV cameras, for example), to make accurate, bespoke recommendations, which can advise travel. The information is presented in the form of interactive maps, news bulletins and a chatbot interface for general queries.

How we built it

For the heatmap, we used google maps API to extract geographical data used to plot the course between two points for various modes of travel. We combined it with the geographical data for COVID statistics extracted from government websites, applying our custom algorithm to it to generate a heat map based on risk. We also trained convolutional neural networks on CCTV footage to detect approximate density of people in an area (such as a mall or a park) and also other specific insights such as the proportion of people with masks on, and the overall level of effective social distancing. The dashboard also contains live updates from local news outlets, which are curated for the user. Additionally, a deep learning-based chatbot has been integrated that offers general advice and guidance.

Challenges we ran into

Most of the members of the team are quite new to app development and machine learning, so those parts of the project were quite challenging. Working with live data also presented some problems. We were also new to using gmaps and the google maps API, which took some time to figure out.

Accomplishments that we're proud of

This project is a prototype for a service that we genuinely believe is important and useful to the everyday user, and it is novel in its approach.

What we learned

In addition to learning how to use new libraries and API, we gained more experience on working together as a team, under pressure, meeting short deadlines.

What's next for COVID Consultant

If time allows, we will add further functionality to the application, allowing more seamless integration into the other services that the user might want to use. We can also spend more time on improving the UI and making the features more intuitive.

Built With

Share this project:

Updates