Date of Submission
Spring 1-18-2021
Degree Type
Thesis
Degree Name
Master of Science in Computer Science (MSCS)
Department
Computer Science
Track
Others
MACHINE LEARNING
Abstract
Diabetes Mellitus (DM) which refers to a metabolic disorder that occurs when the level of blood sugar in the body is considered high, which could be a resulting effect of inadequate availability of insulin in the body. It is a chronic disease which may lead to myriads of complications in the body system. Statistics by the World Health Organization (WHO) in 2013, indicated that DM was the cause of death of over 1.5 million people around the world and in 2016, 8.5% of adults within age seventeen (17) and above were reported to be diabetic and diabetic patients have continued to increase in recent years. It is therefore very glaring that these alarming figures calls for very urgent and effective attention. There has been a recent proliferate increase in studies relating to machine learning in the healthcare sector, hence the motivation for this research work. The research was based on the prevalence of diabetes amongst the masses of Kaduna metropolis using some selected hospitals as a case study after which a predictive model was designed for diabetes, using some selected supervised learning algorithms like Decision tree algorithm, K- Nearest Neighbour algorithm and Artificial Neural Networks on a dataset gotten from 44 Army Reference Hospital and Yusuf Danstoho Memorial Hospital Kaduna which constitutes of nine (9) attributes that was considered. The results indicated that ANN produced the highest accuracy with 97.40% followed by decision tree algorithm with 96.10% accuracy then K-NN algorithm with 88.31% accuracy. This result was further validated using fifty (50) dataset out of which forty-eight results were rightly predicted.
Comments
Keywords
Diabetes mellitus, metabolic disorder, healthcare sector, artificial neural network.