Traditionally, very simple statistical techniques are used in the analysis ofrnepidemiological studies. The predominant technique is logistic regression, in whichrnthe effects predictors are linear. However, because of their simplicity, i.t is difficult tornuse these models to discover unanticipated complex relationships, i.e., non-linearitiesrnin the effect of a predictor or interactions between predictors. Specifically, as thernvolume qj data increases, the traditional methods will become inefficient andrnimpractical. This in turn calls the application of new methods and tools that can help tornsearch large quantities of epidemiological data and to discover new patterns andrnrelationships that are hidden in the data. Recently, to address the problem ofrnidentifying useful information and knowledge to support primary healthcare preventionrnand control activities, health care institutions are employing the data mining approachrnwhich uses more flexible models, such as, neural networks and decision trees, torndiscover unanticipated features from large volumes of data stored in epidemiologicalrndatabases.Particularly, in the developed world, data mining technology has enabled health carerninstitutions to identify and search previously unknown, actionable information fromrnlarge health care databases and to apply it to improve the quality and efficiency ofrnprimary health care prevention and control activities. However, to the knowledge ofrnthe researcher, no health care institution in Ethiopia has used this state of the artrntechnology to support health care decision-making.Thus, this research work has investigated the potential applicability of data miningrntechnology to predict the risk of child mortality based up on community-basedrnepidemiological datasets gathered by the BRHP epidemiological study.rnThe methodology used for this research had three basic steps. These were collectingrnof data, data preparation and model building and testing. The required data wasrnselected and extracted from the ten yea rs surveillance dataset of the BRHPrnepidemiological study. Then, data preparation tasks (such as data transformation,rnderiving of new fields, and handling of missing variables) were undertaken. Neuralrnnetwork and decision tree data mining techniques were employed to build and test thernmodels. Models were built and tested by using a sample dataset of 1100 records ofrnboth alive and Died children.Several neural network and decision tree models were built and tested for theirrnclassification accuracy and many models with encouraging results were obtained. Therntwo data mining methods used in this research work have proved to yield comparablyrnsufficient results for practical use as far as misclassification rates come intornconsideration. However, unlike the neural network models, the results obtained byrnusing the decision tree approach provided simple rules that can be used by nontechnicalrnhealth care professionals to identify cases for which the rule is applicable.In this research work, the researcher has proved that an epidemiological databaserncould be successfully mined to identify public health and sociology-demographicrndeterminants (risk factors) that are associated with infant and child mortality in ruralrncommunities