REAL TIME SOCIAL DATA ANALYSIS FOR FORECASTING NATURAL DISASTERS

Author (s) : Mrs.J.SHANBAGAM & Mrs.M.BRINDHA

---

Abstract :
Online social network billions of users all over the world where people collaborate and share information related to real world events. The status update which almost specifies what is happening around an individual and also around the individual’s location. This small content with real world information when processed with some statistical tool may help us to predict a real world event. Inspired by the concept we propose a method to predict the cyclone formation in India by processing the tweets messages posted by the social users. Our system consist of two levels, first we create a classifier to filter out messages which specify the occurrence of our event, in our case cyclone formation. In second level we apply statistical method over the generated data model to predict the occurrence of the cyclone in future. We also create an alert system for our registered user to deliver a notification during cyclone which helps saving many lives. This online social network is used by millions of people around the world to remain socially connected to their friends, family members. A status update message, called a tweet, is often used as a message to friends and colleagues. A user can follow other users, that user’s followers can read her tweets on a regular basis, many researchers have published their studies of Twitter to date, especially during the past year to create new applications using Twitter. The real time nature of the updates helps follower’s to know about an event. They include social events such as parties, baseball games, and presidential campaigns. They also include disastrous events such as storms, fires, traffic. We propose an event   notification system that monitors tweets and delivers notification promptly using knowledge from the investigation. Many researchers have published their studies of Twitter to date, especially during the past year. Most studies can be classified into one of three groups. Some researchers have sought to analyze the network structure of Twitter; some researchers have specifically examined characteristics of Twitter as a social medium.  Some researchers and developers have tried to create new applications using Twitter by creating much data processing and clustering tools. They are still processing on creating a tool to any these data because of its thin text content. Our system crawl numerous tweets related to target events and use a probabilistic model to extract events from those tweets and estimate locations of event. Finally, we are going to develop an event notification or reporting system that extracts event from Twitter and sends a message to registered users. To obtain tweets on the target event precisely, we apply semantic analysis over a tweet to understand the concept of the posted tweets. We  also  use  these  tweet  information  to  forecast  a  particular  event  by  applying statistical tools over the extracted data model. Our system helps users by sending prior notification than the news or any other media about an event. The forecasting made by the system helps us to get some idea about the occurrence of a particular event in future.

Keywords : Twitter, event detection, social sensor, location estimation, Cyclones.

Conference Date : 17/09/2021

Pages : 39-40

Paper ID : IESMDT38