SN voice and text analysis as a tool for disaster effects estimation — A preliminary exploration

This paper reviews the literature and offers a perspective on speech communication, voice and natural language analysis when speech is used under disaster conditions. The focus is on developing a model for interpreting the number and density (probability) of emotional communications in a social network in disaster areas. The model is based on recent researches on social networks and on evidence reported in the literature on emotions under disaster conditions. Several unexpected conclusions are derived, including that the emotional emotion data picked up at the disaster scene without geographical information is not uniformly distributed in a uniform distribution of the population. In addition, the model shows the geographic distribution of the probabilities of the emotions in communications heavily depend on the population density distribution. Also, it shows how some probabilities of the emotions in communications may be dependent on the number of rescue and mitigation operatives on the site.