Inspiration
Being international students, it's been little tough to keep tabs on what's going on at home / what're the current trends / news locally. Local chirrrp helps us to get local trends in real-time.
What it does
We have a web interface with world map. Upon click on the map, it fetches local trends and shows them in word cloud form.
How I built it
We use Twitter API hosted on Google Cloud to gather local trends with geo filters. We use Google maps and Mapbox APIs for our front-end to fetch geo coordinates. Front end queries server to get local trends in real-time and shows them in word cloud form.
Challenges I ran into
We also wanted to use media sources such as NYT and other credible sources along with Twitter data. However, reliable topic modeling of hundreds of articles on the fly is a little bit difficult and hence we just use Twitter data in this case.
Accomplishments that I'm proud of
The fact that we were able to put up a demo in 12-15 hours is a huge achievement.
What I learned
Google Maps / Twitter / Mapbox APIs, Flask webserver, Google Cloud.
What's next for LocalChirrrp
Currently, we get 5-10 secs of delay when fetching data in real-time due to processing time. We plan to cache global hotspots in advance to speed up the process. In addition, we plan to use offline topic modeling of new article to make our output more reliable.

Log in or sign up for Devpost to join the conversation.