EDIT: A Methodology for the Treatment of Non-authoritative Data in the Reconstruction of Disaster Scenarios

摘要

EDIT - Extracting Disaster Information from Text - is a methodology for the treatment of textual information extracted from the web, originated by non-expert users and referring to the occurrence of a disaster event. The project was born from a collaboration between CIMA Foundation (Italy) and the Nottingham Geospatial Institute (NGI) (United Kingdom). The methodology addresses the task of integrating unstructured knowledge into operative tools. It focuses on: the semantic analysis of online news and tweets; the automatic extraction of relevant data to enhance knowledge about disasters; the evaluation of reliability of data; the archive into an event-oriented database; the visualization of scenario maps. The study represents the starting point of a new way of approaching disaster scenarios and it is proposed that the approach is taken forward for future large-scale implementation.