In this study, ultraviolet-visible (UV-Vis) spectrophotometry method was applied to determine benzotriazole (BTA) in aqueous solutions based on surface plasmon resonance (SPR) property of gold nanoparticles (AuNPs). The experimental variables such as pH, time, the volume of various buffers, the concentration of different electrolytes, and concentrations of AuNPs were investigated. Under optimum conditions, the proposed method represented a linear range between 10 to 150 μg L À 1 (R 2 = 0.9941) with a detection limit (LOD) of 5 μg L À 1 . This method was successfully used to measure trace amounts of BTA in Ahwaz drinking water and Ramin power plant cooling water samples. The proposed spectrophotometry assay is accurate, low-cost, simple, and does not need any separation steps compared to the other methods. In the next step, an artificial neural network (ANN) using Levenberg-Marquardt (LM) and Resilient Back propagation (RP) algorithms, was used to predict the absorbance of the mentioned factors. For each algorithm, 2 layers and for each layer 2,4,6,8, and 10 neurons were studied. According to mean square error (MSE) and coefficient of determination (R 2 ), the LM algorithm was better than the RP algorithm for the prediction of absorbance.[a] Dr.