Robust algorithm of multi-source data analysis for evaluation of social vulnerability in risk assessment tasks

This paper is dedicated to the study of social vulnerability in risk analysis task using multi-source data statistics. Formal KPCA-algorithm for quantitative analysis of social factors of disaster risk is proposed. Modified form of damage function basing on the prospect theory and decision making under uncertainty on cognitive bias and handling of risk is proposed. Formal relations for analysis of damage with age, education and income of population are calculated. Analysis demonstrates that at least half of disaster damage might be caused by social factors: education, long-term experience and social behaviour. Using algorithm and approach proposed it was shown a way to estimate quantitative parameters of the losses distributions in view of dynamics of social drivers.