MindRoad

There are intense automatic data collections and online computing processes on MindRoad backend. In the app only the final results are reflected in a graph format.

At the moment MindRoad only works for NSW because of data collection limitations but can be modified for any other region in the world.

On a daily basis all the tweets posted in Sydney region are automatically scripted from Twitter API and stored in a server. Also, every hour travel times on main corridors in Sydney are automatically collected from Google Maps API and kept on the server.

In the app, when a keyword is entered by a user, MindRoad communicates with the server and filters the tweets that have the entered keyword. Then calls a well established Natural Language Processing (NLP) library and automatically analyses the context of the filtered tweets and scores each of the selected tweets a value between -1 to +1. If a tweet conveys a strong positive opinion receive +1 score and respectively -1 for a strong negative opinion. A tweet might also receive a score between -1 and +1 based on its context.

The summary of all the above-mentioned process is illustrated in a graph on the main page of the app. The graph shows public opinion changes versus average travel times.

This app particularly focuses on Westconnex and NorthConnex projects at the moment and the average travel times for this two projects are more accurate than other keywords.