A framework for monitoring movements of pandemic disease patients based on GPS trajectory datasets

The rapid spread of contagious diseases poses a colossal threat to human existence. Presently, the emergence coronavirus COVID-19 which has rightly been declared global pandemic resulting in so many deaths, confusion as well huge economic losses is challenge. It suggested by World Health Organization (WHO) conjunction with different Government authorities world and non-governmental organizations, that efforts curtail should rely principally on measures such social distancing, identification infected persons, tracing possible contacts effective isolation person(s) for subsequent medical treatment. aim this study provide framework monitoring Movements Pandemic Disease Patients predicting their next geographical locations given recent trend patients absconding from centres evidenced Nigerian case. methodology study, proposes system architecture incorporating GPS (Global Positioning System) Assisted-GPS technologies movements recording movement Trajectory Datasets assumption they are assigned GPS-enabled devices smartphones. Accordingly, fifteen (15) participants (patients) were selected based criteria residency business activity location. ensuing generated 157, 218 datasets during period 3 weeks. With dataset, mining trace, Stay Points (hot spots), relationships, prediction probable location patient was realized application Artificial Intelligence (AI) Data Mining techniques supervised Machine Learning (ML) algorithms (i.e., Multiple Linear Regression (MLR), k-Nearest Neighbor (kNN), Decision Tree (DTR), Random Forest (RFR), Gradient Boosting (GBR), eXtreme regression(XGBR) density-based clustering methods DBSCAN) computation spots) patient. result showed clearly it determine In addition, demonstrated possibility Correspondingly, Six models MLR, kNN, DTR, RFR, GBR, XGBR) compared efficiency, determining DTR model performed better other four evaluation matrices ACCURACY, MAE, MSE, R2) used. recommended less developed Countries consider adopting policy initiative implementation at burgeoning phase infection beyond. same applies Countries. There indication dataset applied paper, appear possess potential performing optimally real-life situation This paper unique its ability predict review extant literature, not evidence using algorithms.