Earthquake Shake Detecting by Data Mining from Social Network Platforms

Yu-Jung Chuo

doi:10.3390/app10030812

Abstract

This study used social media posts of the related effect of earthquakes to derive seismic shake scale distributions in regions of Taiwan and compared it with the regional seismic scale reported by the Central Weather Bureau (CWB) of Taiwan. This study conducted a context searching to scrawl the relationship phrase on the social media network platform, PTT bulletin board system (BBS), to detect the earthquake shake scale using the keywords of the context. In this investigation a decision tree model for analyzing the semantic words from the context of the target event to detect the earthquake shake scale was devised. The results indicate that we can pick out the keywords to use to detect the earthquake shake scale at about 85%. Furthermore, the results of the derived shake scale show that the four studied cases are in a good agreement with the presented news from the CWB of Taiwan. In this study, the author attempted to develop a quick earthquake shake scale detection model by semantic analysis of the collected earthquake disaster information reported on the social media network platform.

Highlights

With the increasing development of network and wireless communication, the social media network has become a platform for people to vent their emotion and feelings in real time and stores a large amount of information
Shen et al [11], reported that the data of the social media network increases daily, and the data can be offered for studying, training, and testing on different issues by using the data mining method. They collected some posts as examples from the bulletin board system in Taiwan, PTT, to develop a real- time depression detection system to identify potential depression candidates based on their writing
This study examines the investigation by using the developed decision model withoccurrence, other earthquakes collected from posted the PTTtoBBS

Summary

Introduction

With the increasing development of network and wireless communication, the social media network has become a platform for people to vent their emotion and feelings in real time and stores a large amount of information. Shen et al [11], reported that the data of the social media network increases daily, and the data can be offered for studying, training, and testing on different issues by using the data mining method. They collected some posts as examples from the bulletin board system in Taiwan, PTT, to develop a real- time depression detection system to identify potential depression candidates based on their writing.

Methods

Results

Conclusion