Prediction of Surface roughness plays a vital role in manufacturing process. In manufacturing industries, productions of metallic materials require high surface finish in various components. In the present work, the effect of spindle speed, feed rate, depth of cut and flank wear of the tool on the surface roughness has been studied. Carbide tipped insert was used for boring operation. Experiments were conducted in CNC lathe. The experimental setup was prepared with sixteen levels of cutting parameters and was conducted with two tool tip conditions in dry machining. A piezoelectric accelerometer was used to measure the vibrational signals while machining. The data acquisition card which connected between accelerometer and lab-view software to record the signals. Simple linear and least median regression models were used for prediction of surface roughness. The models were developed by weka analysis software. The best suitable regression model is implemented based on maximum correlation coefficient and the minimum error values.