Developing Knowledge Based System for Predicting HIV/AIDS Stages Using Data Mining Techniques

Abstract

The Human Immunodeficiency Virus (HIV) is an infectious disease that attacks the immune system cells. HIV disease is the most killer disease among humankind, especially in adult age level. Data mining is the process of extracting previous unknown patterns and knowledge to design predictive models from large patient data. The huge amounts of patient data generated for the prediction of HIV Stages are time-consuming and too complex to processed and analyzed by manual methods. Prediction of HIV/AIDS Stages is a major problem using traditional method faced in hospitals and other clinical centers. The researcher is initiated to develop Knowledge-Based System and identify HIV/AIDS patient’s stages using different data mining technique based on the data which is collected from the hospitals. The researcher use Knowledge Discovery in Databases Data Mining methodology for this research study. To identify the stage of HIV patient the researcher used three data mining classification algorithms; J48

Decision tree as well as, PART and JRip rule induction. The PART classification algorithm is selected for use in the development of a knowledge base because it registered the best accuracy of 98.4885 %. The evaluation result of the proposed knowledge-based system achieves a promising result of 89.7959 % accuracy and 76% user acceptance testing. To accomplish this study, the researcher needs an accurate dataset.