With its ability to promote a more environmentally sustainable future, biodiesel, a renewable fuel made from plant oils and animal fats, shows great promise. However, costly, ineffective, and ecologically harmful homogenous catalysts are challenges that conventional production processes face. Geopolymer catalysts and machine learning approaches are starting to show promise as game-changers in the biodiesel industry. Three machine learning algorithms, response surface methodology (RSM), artificial neural network (ANN), and adaptive neuro-fuzzy inference system (ANFIS), were used to optimise and model biodiesel production from waste cooking oil using process parameters, such as methanol-to-oil ratio, catalyst loading, reaction temperature, and reaction time. RSM was used for process optimisation. The model construction of the ANN model used 70% of the data for training, 15% for testing, and 15% for validation. The network was trained using feed-forward propagation and the Levenberg--Marquardt algorithm. The ANFIS was generated using a grid partition and trained using a hybrid method. The effectiveness of the machine learning was assessed through error metrics, such as regression (R²), root mean square errors (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), average relative errors (AREs), and mean percent standard deviation (MPSD). The optimum yield was obtained witha methanol-to-oil ratio of 15 wt. %, a catalyst loading rate of 4 wt. %, a temperature of 120^o C, and a reaction time of 4 h, yielding 93.486 %. The results have shown the promising use of machine learning potential methods for optimising, modelling, and predicting the methanolysis of waste cooking oil using geopolymers for eco-friendly biodiesel production.