Loading [MathJax]/extensions/MathZoom.js
Twitter Analysis for Disaster Management | IEEE Conference Publication | IEEE Xplore

Twitter Analysis for Disaster Management

Publisher: IEEE

Abstract:

Disaster always leads to response from the people and in this new technology generation, people give their feedback and raise concerns through their views on social media...View more

Abstract:

Disaster always leads to response from the people and in this new technology generation, people give their feedback and raise concerns through their views on social media platform. Twitter is used for such response and information exchange. This paper reports in depth twitter analysis of Nepal Earthquake and identification of various factors in the disaster module. Based on Twitter data collected between last week of April and first week of May 2015. 40,236 raw messages apparently related to the Nepal earthquake were retrieved, preprocessed and analysis were done on the same. This paper shows measures and analysis of the situation using geolocation feature for identifying danger zones and visual analytics of the dataset. With the help of the keyword generation algorithm different parameters were introduced. The results show that our disaster module can work on different hashtags and no need for prior defining of parameters by humans provided the data set is available for that unique problem.
Date of Conference: 16-18 August 2018
Date Added to IEEE Xplore: 25 April 2019
ISBN Information:
Publisher: IEEE
Conference Location: Pune, India

I. Introduction

We analyze the contents of Tweets regarding a particular Hashtag and develop a detailed analysis of the contents and mapping the co-ordinates of the users using Google maps with intent to gain a broad understanding of the nature of concerns and communication. We have used a machine Learning algorithm to automatically generate the keywords based on the data fed and then generate a detailed analysis upon those keywords while also mapping the geo-location co-ordinates of users, categorizing the users as ‘In danger’ and one's ‘providing help’ and also adding their respective tweets along the username on the map and thus creating a Danger map with twitter data

References

References is not available for this document.