CrisMap: A Big Data Crisis Mapping System Based on Damage Detection and Geoparsing
| Autor: | Stefano Cresci, Fabio Del Vigna, Tiziano Fagni, Marco Avvenuti, Maurizio Tesconi |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2018 |
| Předmět: |
Situation awareness
Computer Networks and Communications Computer science Big data 02 engineering and technology Theoretical Computer Science Geoparsing Social media 020204 information systems Benchmark (surveying) 0202 electrical engineering electronic engineering information engineering Choropleth map Natural disaster Online social networks Crisis mapping business.industry Crisis mapping · word embeddings · geoparsing · online social networks · social media · big data Data science Word embeddings 020201 artificial intelligence & image processing business Software Information Systems |
| Zdroj: | Information systems frontiers (Dordrecht. Online) 20 (2018): 993–1011. doi:10.1007/s10796-018-9833-z info:cnr-pdr/source/autori:Avvenuti M.; Cresci S.; Del Vigna F.; Fagni T.; Tesconi M./titolo:CrisMap: a Big Data Crisis Mapping System Based on Damage Detection and Geoparsing/doi:10.1007%2Fs10796-018-9833-z/rivista:Information systems frontiers (Dordrecht. Online)/anno:2018/pagina_da:993/pagina_a:1011/intervallo_pagine:993–1011/volume:20 |
| DOI: | 10.1007/s10796-018-9833-z |
| Popis: | Natural disasters, as well as human-made disasters, can have a deep impact on wide geographic areas, and emergency responders can benefit from the early estimation of emergency consequences. This work presents CrisMap, a Big Data crisis mapping system capable of quickly collecting and analyzing social media data. CrisMap extracts potential crisis-related actionable information from tweets by adopting a classification technique based on word embeddings and by exploiting a combination of readily-available semantic annotators to geoparse tweets. The enriched tweets are then visualized in customizable, Web-based dashboards, also leveraging ad-hoc quantitative visualizations like choropleth maps. The maps produced by our system help to estimate the impact of the emergency in its early phases, to identify areas that have been severely struck, and to acquire a greater situational awareness. We extensively benchmark the performance of our system on two Italian natural disasters by validating our maps against authoritative data. Finally, we perform a qualitative case-study on a recent devastating earthquake occurred in Central Italy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|