Research Article

Predicting Social Unrest Events with Hidden Markov Models Using GDELT

Figure 3

The proposed HMMs-based social unrest event prediction framework: two HMMs are trained, with one for -prone sequences and one for -free sequences. -prone sequences consist of observations within a time window of length preceding a social unrest event (▲) by lead time . -free sequences consist of observations at times when no social unrest event was imminent. is the time the prediction is performed at. is the prediction period.