Discover Your Island University

Graduate Projects


Project ID: 439
Author: Nikhil Srinivas Komatireddy
Project Title: Disaster Detection and Reporting System through Analysis of Twitter Tweets
Semester: 2 2015
Committe Chair: Dr. Longzhuang Li
Committee Member 1: Dr. David Thomas
Project Description: Real time events that are described as natural disasters can be investigated in Twitter and the tweets are monitored to detect a target event like a hurricane. An event in Twitter is identified as a target event by classifying the tweets based upon some of the key parameters like the keywords in a tweet, the number of words and their context. A large number of tweets are scanned wherein these target events are involved and are stored into a database. Then the events that are of our interest only are extracted from all the data that is saved and the locations of these events are estimated. All the registered users of the notification system get a notification through their registered email IDs whenever a hurricane occurs. These valid customers can keep themselves and their loved ones safe by always being informed about this dangerous life-taking event, which occur prior to no information. This notification system is even more helpful when there is a higher risk of getting trapped where there is no availability of television or other sources of information. Unlike other reporting systems, this project can distinguish between a storm, a hurricane and a tsunami based on the key characteristics like the wind speed, rainfall, rise in sea level and time period. It just simply does not notify the registered users without checking for trustworthiness. It only notifies if the message in the tweet is true. This project can be improved by sending message alerts through the telecommunication network companies.
Project URL:   439.pdf