A tree based neural network for forecasting outpatient health-care services demand – A Case Study of medical facilities Nairobi County, Kenya

Abstract

Machine learning models are increasingly gaining wide applications in different fields with an emphasis on supervised machine learning for data exploration in health-care provision. Increase in the demand for outpatient health services has made it difficult to forecast the demand for outpatient services in Kenya. It is in this regard that this research developed and evaluated a tree based artificial neural network for forecasting the demand for outpatient health- care services in Nairobi County, Kenya. Data used in this research was secondary data obtained from the different medical facilities in Nairobi. Experimental results showed that age and gender of an outpatient were significant factors in the estimation and forecasting of demand for outpatient health-care services as they had respective 0.4134 and 0.5646 unit decrease on demand for outpatient health care for a unit increase in demand. Furthermore, the neural network architecture used for forecasting was seen to perform better in forecasting demand for outpatient health-care services. Keywords: Supervised Machine Learning, Outpatient health-care Demand, Artificial Neural network