KNN (K-nearest neighbor) for G-LOC Prediction

G-stress can cause a reduction in the consciousness of jet fighter pilots due to G-induced loss of consciousness (G-LOC). G-LOC is a severe problem for aircraft pilot. We are now trying to use machine learning to investigate the prediction of G-LOC/A-LOC. This research uses machine-learned computational analyses to predict the cognitive performance impairment of the volunteers induced by high gravity. We selected 250 samples for the machine learning. After the research, we apply a pentafold K-Nearest Neighbor (KNN) model to do the supervised machine learning, and we got the G-LOC prediction model with the precision of 85%, which is eligible to predict the status of the pilot before they do the human centrifuge training.