Shear strength of a soil is perhaps the most important of its engineering properties, as stabilityrnanalyses in the field of geotechnical engineering are dependent on it. This research work seeks torndevelop models for predicting the shear strength parameters (cohesion and angle of friction) ofrnsoils in Addis Ababa city using artificial neural network modeling technique; with a view tornreducing time, effort and cost usually incurred in determining these shear strength parameters inrnthe laboratory for future planning, design and construction projects in the study area. An attemptrnhas been made to develop separate neural network models for c and Ï• from the index properties ofrnsoil consisting of Sand % (SP), Fines % (FP), Liquid limit (LL), Plasticity Index (PI), water contentrn(Ï‰), and Bulk density (BD) as input parameters. A multi-layer perceptron network with feedrnforward back propagation is used to model varying the number of hidden layers. For this purpose,rn284 soil test result data was used. The geotechnical soil properties were determined in accordancernwith ASTM Standards. Direct shear box method was used to determine soil cohesion and soilrninternal friction angle. The developed models were found to be quite satisfactory in predictingrnshear strength parameters with correlation coefficients of about 0.98 and 0.92 for cohesion andrnangle of internal friction, respectively during the testing phase. The models are validated byrnprimary soil test data and compared with some existing correlation methods. The result showedrnthat the artificial neural network method gave better fit and accuracy than the selected empiricalrnformulae in the prediction of shear strength parameters.