An experimental mixture design coupled with data analysis by means of both response surface methodology(RSM) and artificial neural network(ANN) was applied to explore the optimum process parameters for adenosine extraction from cultured mycelium of Cordyceps militaris. With the extraction rate of adenosine as index, the critical factors selected for the investigation were extracting temperature, extracting time and solid-liquid radio. The results obtained by the application of ANN were more reliable since better statistical parameters were obtained. The optimum extraction procedure was as follow: extracting time 2.3 h, extracting temperature 48 °C, solid-liquid ratio 1:38 g⋅mL-1. Under the optimal conditions, the corresponding response value predicted for adenosine production was 4.59 mg g-1, which was confirmed by validation experiments.