Prediction of HIV Status from Demographic Data Using Neural Networks

Abstract

Neural networks are used as pattern recognition tools in data mining to classify HIV status of individuals based on demographic and socio-economic characteristics. The data consists of sero prevalence survey information and contains variables such as age, education, location, race, parity and gravidity. The radial basis function (RBF) neural network architecture was used for this study since as preliminary design showed this architecture to be the most optimal. The Bayesian method of training used was approximated with the evidence framework. The design of classifiers involves the assessment of classification performance, and this is based on the accuracy of the prediction using the confusion matrix. An accuracy of 84.24% was obtained in this design. This thus implies that the HIV status of an individual can be predicted using demographic data to 84.24% accuracy. A network comprising of 9 primary RBF, and MLP networks of structure 1-3-1 (input-hidden node-output node) and one secondary MLP network of structure 9-77-1, was used with a prior of 0.24693 and 144 training cycles which was found as the optimal training cycles.