Predicting and modeling flank wear length in high speed hard turning by using ceramic cutting tools with negative rake angle was conducted using two different techniques. Regression model is developed by using design of expert 7.1.6 and neural network technique model was built by using matlab2009b. A set of experimental data for high speed hard turning of hardened AISI 4340 steel was obtained with different cutting speeds, feed rate and negative rake angle. Flank wear length was measured to train the neural network models and to develop mathematical model by using regression analysis. Predictive neural network models are found to be capable of better predictions tool flank wear within the range that they had been trained.