Anomalous Gait Detection using Naive Bayes Classifier

The aim of this study is to investigate the potential of Naive Bayes classifier as abnormal gait pattern detection specifically due to Parkinson Disease since it is vital to identify the best classifier that can perform competitively prior to implementation of a gait identification system. Moreover, the significant of SFS short for 'sequential feature selection' is experimental explored along with Naive Bayes capability as classifier. Initial findings showed that classification task based on Naive Bayes is extremely competitive based on the highest accuracy rate attained specifically 93.75% through sequential feature selection and 84.38% otherwise. This finding confirmed that Naive Bayes precisely with SFS is among the most suitable classifier for detection of abnormal gait pattern in PD.