Sensory evaluation is widely used in the food industry, especially for quality inspection, product design and marketing. Ice cream acceptability by consumers is mainly perceived by means of texture and flavor. Various ingredients contribute to ice cream's complex colloidal structure and make its perception difficult. In addition, it requires training the judges and proper environmental conditions during assessment. Furthermore, one distinctive characteristic of sensory responses is that they are ambiguous and imprecise; that is, they are fuzzy. Then, by normal statistical analysis of sensory data obtained through subjective evaluation, a complex idea of a product quality is often generated, which makes it nearly impossible to determine the strength and weakness of the product concerning its sensory attributes. Therefore, application of fuzzy set concept could be useful in this area. Combination of fuzzy concept and artificial neural network is of great importance among various combinations of methodologies in soft computing. There are some reports dealing with the sensory evaluation of ice cream based on texture perception. However, there are no or few studies available concerning the use of computing technology for prediction of consumer acceptance of ice cream. Hence, the objectives of this work were to investigate the efficiency of genetic algorithm–artificial neural network (GA-ANN) and adaptive neuro-fuzzy inference system (ANFIS) simulations for predicting acceptability of hard ice creams during storage. For this purpose, GA-ANN and ANFIS were fed with 4 inputs of flavor, iciness, wateriness and creaminess for prediction of overall acceptability of ice cream. Both models were trained with experimental data. The developed GA–ANN, which included 16 hidden neurons, could predict total acceptance with correlation coefficient of 0.93. The overall agreement between ANFIS predictions and experimental data was also very good (r = 0.92). Results of present research showed that both GA-ANN and ANFIS models’ predictions agreed well with testing data sets and could be useful for understanding and controlling factors affecting palatability of ice cream.