查看文章

To visualize post-emergency damage, a crisis-mapping system uses readily available semantic annotators, a machine-learning classifier to analyze relevant tweets, and interactive maps to rank extracted situational information. The system was validated against data from two recent disasters in Italy.