Water contamination due to the artificial dyes from domestic and industrial wastewater is one of the most crucial environmental issues and problems that the use of an effective adsorbent for dye adsorption seems necessary. In the present study, calcium alginate hydrogels reinforced with cellulose nanocrystals (CA/CNC) as a green and cost-effective adsorbent were utilized for the adsorption of methylene blue (MB). The characterization of CA/CNC hydrogel beads was performed by Fourier transform infrared spectroscopy (FTIR), scanning electron microscopy (SEM), and Brunauer–Emmett–Teller (BET). The MB adsorption kinetics data were consistent with pseudo-first-order model (R2 > 0.99). The equilibrium data of MB adsorption was best fitted by the Langmuir isotherm model and showed monolayer adsorption on homogeneous sites. The maximum Langmuir adsorption capacity (qmax) as an index of the adsorption performance was obtained as 676.7 mg g−1. The obtained results were analyzed by response surface methodology (RSM) and artificial neural network integrated with the whale optimization algorithm (ANN-WOA). ANN-WOA was coded using Python programming language. The contact time (min), shaking rate (rpm), and MB dye concentration (ppm) were considered as input factors for both methods. The fit of the predictive model by RSM was good enough with a correlation coefficient of 0.987. The ANN-WOA model with 3:7:1 topology resulted in higher correlation coefficient, lower root mean square error, and lower normalized standard deviation of 0.999, 0.758, and 15.320, respectively. However, evaluating the statistical criteria confirmed that ANN-WOA is superior to RSM for predicting the experimental data. Therefore, this work showed that CA/CNC hydrogels can be considered as a bio-adsorbent with a simple fabrication route and good adsorption capacity to remove MB from contaminated waters.