Detection Of Urban Emergency Events Using Social Media









Abstract

Events occurring in urban areas e.g. earthquakes, blasts, traffic jams, crimes, fires etc. pose a huge threat on the lives of people. These events are very important to be monitored to make an assessment and prevent any damage occurring to life and property. In order to do the monitoring and the detection of such events the 5W (What, Where, When, Why, and Who) model is proposed. Crowdsourcing is the process to acquire, integrate and process the enormous data generated by sensors, devices and humans. In this paper the proposed model is using the social media as a platform to detect the emergency events and aware them for the same. Through user’s social media post our model would be able to gather the information required and produce the result accordingly and then necessary steps will be taken for the betterment of surrounding. First of all, users on the social media are made sensors through which crowdsourcing (WHO, WHAT) occurs. Secondly, the geo-spatial and terrestrial information are extracted from the social media for determining the real time events (WHERE, WHEN). Thirdly, an annotation of GIS (Geographic Information System) of the detected emergency event is shown. The data is collected and forwarded to higher authorities approved by system admin to take necessary action. Keywords - Emergency events, Crowdsourcing, Social Media, Data Mining.


Modules


Algorithms


Software And Hardware

• Hardware: Processor: i3 ,i5 RAM: 4GB Hard disk: 16 GB • Software: operating System : Windws2000/XP/7/8/10 Anaconda,jupyter,spyder,flask Frontend :-python Backend:- MYSQL