Twheat Map

twitter

General description: A real-time map of tweets from all over the world.

Contributors: Kazem Jahanbakhsh, Ajay Promodh Sridharan, Tim Song, and Priyanka Gupta

Features: Find the sentiment analysis of geotagged tweets in real time.

Implementation period: September 28-29, 2012

Technical details:
This application was developed during a 24-hour Abebooks Hackaton event in Victoria [1]. In this hackathon we wanted to build a cool web application using the Amazon Web Services (AWS) cloud computing platform [2]. Our main idea was to collect twitter public stream [3] and then run machine learning algorithms on top of that for analyzing tweets. This analysis allows us to better understand what is going on in the world in real time as well as people opinions around different topics. In this competition we used a "Naive Bayes" classifier to do sentiment analysis on tweets to find out if they are positive or negative.

Analyzing tweets has several applications in different domains such as predicting the future (e.g. predicting election results) or computing people opinion about different products such as smartphones (e.g. for market advertisement). One of our early results for US 2012 presidency election by counting tweets were showing that Obama campaign is doing better job than Romney's where we have found that around 60% of politic discussions in Twitter was about Barack Obama. We have plan to puruse our research to get better insight about people political opinions by using twitter. This might allow us to test the possibility of predicting the result of upcoming US presidency election in November!

The following figure shows a snapshot of our twheat map where we displayed geotagged tweets with different colors. Here, "red" color shows a positive tweet whereas "blue" color represents a negative tweet ("green" is neutral). Having this real-time map we can find out the mood of different cities in the world in real-time.

sentiment analysis of twitter



A Demo Video from AbeBooks Hackathon 2012


Our team won 300$ prize in this competition :)

Me and Ajay in Abebooks hack day!

Abebooks Hack Day!

Refs:
[1] Abebooks Hackathon
[2] Amazon Web Service
[3] Twitter Public Stream



Tweet

You should follow Follow @kjahanbakhsh me on Twitter.