Skip to main content
Fig. 4 | BMC Psychiatry

Fig. 4

From: Recognizing states of psychological vulnerability to suicidal behavior: a Bayesian network of artificial intelligence applied to a clinical sample

Fig. 4

a The figure displays a Principal Component Analysis performed with the DEQ variables, selecting the first two principal components. The first graph shows the distribution of the projected points in these two components. The second graph shows the contribution of each of the DEQ questions to the first two principal components. b Shows classification error obtained in the model calibration process, using different search algorithms. The Tabu search algorithm was selected due to its lower average classification error and lower variance in the cross-validation sample

Back to article page