H2S removal from an off-gas stream was performed in a spray column by H2S reactive absorption into a NaOH solution. The individual and interactive effects of three independent operating variables on the percentage of absorbed H2S were investigated: the initial pH of the scrubbing solution, the initial scrubbing solution temperature, and the volumetric liquid-to-gas ratio. The optimum operating variables were determined by response surface methodology (RSM) attaining a percentage of absorbed H2S of 98.7 +/- 0.2 %. Additionally, the process performance was modeled by an artificial neural network (ANN) to predict the percentage of absorbed H2S. The results showed that the experimental data agreed better with the ANN model than with the RSM results.