The commercial banking sector plays vital role in the economic developmel1l efforts a/a country.rnAnd the viability of the sector relies on the alibi(v of the institutions to maintain a positive inflowrnof resources. And one of the factors that affect the ability of the commercial banking sectors tornmaintain positive flow of resources is the problem of default rates. Among the Ethiopian bankingrnsector there is a general problem of high default rate and the commercial banks have tried torntackle this problem through different ways.rnOne technique that has become popular in addressing problem of credit risk in other countries isrndata mining. Data mining technology has enabled banks in other countries to make goodrnprediction on the probability that a certain borrower would default or not. BUI, so jar, norncommercial bank in Ethiopia has used data mining technology for such purposes i.e. assessmentrnof credit risk. l1ws, the objective of this research work was to serif application of data miningrncould also be beneficial in the Ethiopian banking context. For reasons of familiarly D ashen BankrnS. C was selected as a case study.rnThe methodology employed for the research handbag Sic ally three stages. These were collection ofrndata, preparation of data and model building and testing. Data was collected hi-om two kinds ofrndocuments that were available at the head office of D ashen Bank. 7'l1Cn the data was preparedrnwhich included stigmatization, deriving of minefields and handling of missing data. The datarnmining technique employed for the model building and testing was neural network. Neuralrnnetwork software was thus used in building and test inga number of models From the numerous trials, many models with encouraging results were obtained. These modelsrnindicated that data mining application for credit decision-making is feasible at D ashen Bank. Butrnone major limitation was unavailability of data in an electronic form. However, a survey in the ITrndeportment of D ashen Bonk suggested that this problem is being duly addressed.