HIV/AIDS has claimed the lives of millions and has left behind hundreds of thousands of orphans in Ethiopia. The government of Ethiopia took several steps in preventing further disease spread, and in increasing accessibility to HIV care, treatment and support for persons living with HIV. Antiretroviral therapy (ART) is one of the treatments given to HIV patients Felege Hiwot Referral Hospital (FHRH) to restore patients with severe disease to healthy.

The dataset for the study contains pre ART records of the year 2005 and 2006 E.C produced by the ART office of patients Felege Hiwot Referral Hospital. The dataset has been utilized for the purpose of predicting clients' eligibility for ART.

Before these data has been used for the purpose of classification a number of pre-processing steps such as data cleaning, data reduction and data transformation have been effectively used which helped in achieving the objective finally or to increase the speed and efficiency of mining process.

The final goal of this paper is to build ART eligibility predictive model that helps to deciding whether HIV positive individual should start Anti-retroviral treatment or not. For building ART eligibility predictive model, Naive Bayesian Classifier and J48 Decision Tree Classifier are used.

After experimenting J48 decision tree and Naive Bayesian classifier using both 10-fold cross validation and percentage split (66%) test modes, J48 classifier using 10-fold cross validation that performs well and can be used as a best predicting model algorithm than Naive Bayesian classifier in predicting clients' eligibility for ART is created.