The unconfined compressive strength (UCS) of cement-stabilized soil depends on the particle size distribution, the soil chemistry composition, the cement content, the water content, and the curing time. Determining the UCS of soil-cement material is an important task in the design and construction of infrastructure supported on marginal soils that require ground improvement. In this research, the Artificial Neural Network (ANN) technique was applied to generate the UCS prediction model in relation to the selected parameters. A sensitivity analysis was conducted to examine the range of the contribution of variables. As a result, a statistical analysis shows that the proposed model developed in this study is accurate and reliable with a high correlation coefficient and low root mean squared errors. The UCS prediction model also satisfies well the external criteria; hence it illustrates the great potential of the predictive ability of the ANN technique for geotechnical parameters. The ANN-based model shows that the cement content and the percentage of the soil particle passing sieve 0.5 mm are the most important variables affecting the UCS value. The research results could be used for estimating the UCS of cement-stabilized sandy soil in situ and laboratory conditions.