Vaccine Distribution Network

Inspiration

Vaccines are an essential and integral part of our medical ecosystem. During COVID-19 when the vaccines were initially released, they received heavy criticism from specific communities due to misinformation. Yet, governments tried to push these critical and perishable vaccines into these communities which caused a lot of social unrest.

What it does

The implemented solution tries to implement a Vaccine distribution system that considers people's sentiments and creates a geospatial cluster map of opinions that the governments could use to efficiently distribute the vaccines in regions where people would like to have them.

Meanwhile, through community education, the rest of the communities could be educated, and they can get the vaccine later. This helps distribute the vaccines efficiently considering the people's opinions.

How we built it

Using the data collected from the Twitter API, we built an application that uses unsupervised machine learning to cluster the data and classify the sentiments. Then a special API is used to reverse geolocate the location and a mapping API is used to generate geospatial cluster maps of the inference from the ML model.

Challenges we ran into

As the Twitter data had a lot of special characters, emojis and mixed sentiments cleaning and designing the mode were the biggest challenges.

Accomplishments that we're proud of

We were successfully able to clean most of the text data using special filters we created.

What we learned

We learned that by using proper text processing schemes we can significantly improve the ML models' accuracy.

What's next for Vaccine Distribution Network

We plan to implement a real-time streaming interface that could perform the ML inference in real time and display the data in real time on a dashboard.

Built With

• geopy
• python
• sklearn