Sirius-1110 / traffic-congestion-prediction

Proposed Proof of Concept to solve Traffic Congestion and Prediction Problem for Smart Cities by using LSTM(for future prediction) + CNN(To detect traffic density) + using tweets and combine all three to produce a traffic congestion factor and suggest routes based on that using Google API at any time in future.

Traffic Prediction

Proposed Proof of Concept to solve Traffic Congestion and Prediction Problem for Smart Cities. Used - LSTM(for future prediction) + CNN(To detect traffic density) + Real Time Tweets Combined all three to produce a traffic congestion factor(TCF) and suggest routes based on that using Google API at any time in future.

Contents -

Public Folder -

TCF Data till 18th September

LSTM Predicted Time till 18th September

src Folder - React Application

Utility Files -

tweets_realtime.py - To scrape geolocation tagged tweets about traffic

tempserver - Temporary server to queue all tweets recieved from scraping and act as an API for the React Application.

Screenshots

Note: Code Cleanup is still in progress, this project was made for a hackathon