Abstract
Terrorism adversely impacts the investment decisions of multinational enterprises. This research proposes the iterative outlier analysis heuristic capable of utilizing a univariate attribute to extract high impact attacks (HIA) from a comprehensive and actively maintained global terrorism database (GTD). The two different univariate attributes used for HIA extraction were nkill (pure GTD univariate attribute) and global terrorism impact score (GTI-IS) (a derived composite attribute). Calendar year extraction of HIAs returns local point outliers, whereas taking whole dataset at once provides global point outliers. The nkill-based extraction resulted in 5 and 14 times more exclusive HIAs than GTI-IS in local and global point analysis, thereby establishing its superiority. Further, correspondence analysis using extracted HIAs demonstrated the affinity towards explosives in the Middle East & North Africa region, with Military and Business as preferred targets. Also, HIAs facilitated the geospatial visualization of terrorism hotbeds. Eliciting location-specific relationships from HIAs on weapon type and target type can assist in formulating better counterterrorism strategies. This study scrutinized the averaging out evaluation methodology of GTI ranking based on GTI-IS score. By equalizing terrorist attacks of distinct outcomes using an average score can induce bias among business decision-makers interested in a particular nation.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
