Earthquake Reporting System Development by Tweet Analysis with Approach Earthquake Alarm Systems

Natural disasters like Tornadoes, floods and earthquakes are only a few of the catastrophic events which may have devastating repercussions over vast geographical regions. Social media is now being identified as one new information source from common public for helping attain productive social outcomes. A few examples for this are monitoring any ongoing disaster incidents, tracking opinion of public, marketing, research human behavior along with public health problems and identifying earthquakes. Considering the huge volume of data that is available on a number of platforms pertaining to social media presently in usage, one notable challenge will be extracting valid and related data toward these various objectives. A targeted and quick response toward emergencies will contribute greatly in reducing the losses caused. The earthquake grid analysis project performs one crowd-origin earthquake advance warning system on the basis of tweet information from users and a warning system procedure via admin server. The said study puts forward a statistical technique toward diagnosis of earthquakes out of the information that comes from a grid of tweet data by the twitter dataset. Also in this study, it is suggested to tweet the earthquake-related data analysis by employing semantic clustering and Nave Bays with filtering process. Gather tweet information from twitter about earthquake first and pre-process it for reducing noise using null values for removing the unnecessary tweet data. Here, we have suggested the process of Gaussian filter toward filtering incorrect user data and processing the accurate information about earthquake. In addition, our technique permits maintaining the false alarm probability under check. The statistical method gets applied to information gathered by position-based earthquake data in the twitter dataset. Also in this study, we have introduced two primary procedures, namely, admin and the user process related to earthquake tweet data handling and alert data forwarding to safe region. This analysis processes four classifications: filtering, preprocessing, Nave Byes with the semantic cluster. Results from experiments prove that the suggested system is highly qualified, meaning that both correct information recovery and distortion diagnosis are being successfully performed.