Abstract

Diagnosis of motor learning difficulties refers to the identification of motor difficulty and rehabilitation means the ways of restoring to its default function. This paper explains how virtual reality can be innovatively integrated to perform this process effectively by leveraging machine learning. Identifying one’s motor learning difficulties often requires evaluation by an expert. Rehabilitating one motor problems also requires supervision. The paper explains the methodology of automating the task of identification and rehabilitation of motor learning difficulty. Results and findings from an early experiment suggest how this process can be integrated into the healthcare system in conjunction with the traditional methods. It also explains the consistency, reliability, and accuracy of this newly developed technology. The ultimate aim of this research paper is to propose a new paradigm shift in the way we diagnose and rehabilitate people with motor learning difficulties.
Full Paper

Abstract

Diabetes Mellitus, known popularly as Diabetes, is a metabolic disease that results in the rise of glucose levels in ones body. Diabetes is influenced by factors such as Insulin, Age and Blood Pressure. In this study, using the Pima Indians Diabetes Dataset, Logistic Regression is performed to predict the possibility of occurrence of Diabetes. Logistic Regression is implemented when the dependent variable(s) are categorical instead of numerical. The model used here is a Logit model which is obtained by performing logarithmic operations upon the Logistic model. A rigorous procedure of data cleaning and outlier elimination results in a model with high accuracy. A logistic model, which is classically used for binary dependent variables, is developed. Confusion matrices assist in the determination of model accuracy. Varying the number of parameters taken into consideration, the effect on AIC is noted. The models performance is further examined by implementing three techniques : AUC and ROC, KS statistics and Gain chart, and finally, Lift test. Percentage change in outcome is determined for each of the independent variables. The gains chart shows that 100% of the target values are covered by the first 80% of the target values alone. the lift chart depicts that 70% of the model records, account for 2.7 time the total targets found by selecting 70% of a file that doesn’t have a model.
Full Paper

Abstract

Power system is a grid of electrical apparatuses used to supply and transmit power [1].
Telecommunication is the use of telephone to send traffic from a source to destination through transmission media.
A happy customer is a returning customer. When customers are not satisfied with the network services they pay for or do not get value for their money, they will not want to patronize such services. In this research, we will design and develop a web-based software call digitalize power system in telecommunication network environment. In this research, our approach to the research will be qualitative and interview as a research instrument will be adopted in collecting data. A prototype model will be applied in developing the system. The system will periodically send Short Message Service (SMS) notification on power status at cell sites, minimize the transportation fare to cell sites and boost revenues from such cell sites.
Full Paper