This paper illustrates the application of the artificial neural network for adsorption of ammonium NH4 + and COD from fish farm by rice straw as low cost carbonaceous. The effects of input parameters (contact time, pH, initial concentration of NH4 + and COD, adsorbent dosages, and temperature) are studied to optimize the conditions for maximum removal of NH4 + and COD. The artificial neural network with a single hidden layer with ten nodes trained with Levenberg-Marquardt algorithm predicted the removal efficiency of NH4 + and COD from aqueous solution accurately.