Predicting Perpetrators of Terrorist Attacks
This mini project uses the global terrorism database (GTD) information on known perpetrators of terrorist attacks, to predict the likely perpetrators for incidents where the group resposinble is not known.
The GTD is an open-source database including information on terrorist events around the world from 1970 through 2014, in which some portion of the attacks have not been attributed to a particular terrorist group.
This model uses a random forest ensemble trained on known groups to predict which groups may be responsible where the group is unknown. Features included the attack type, and weapons used.