Precipitation is the most important weather parameter, which has direct and indirect effects on life on our planet.  One of the most relevant climate change consequences is extreme weather conditions, such as floods and droughts.  Up to now, climate model projections have remained uncertain, therefore, more accurate precipitation modelling techniques are necessary. Due to the complexity of relationships among metrological parameters, traditional statistical modelling tools are ineffective in forecasting and predicting weather conditions. In this study, data mining techniques were applied to metrological data to predict daily precipitation using Multilinear regression (MLR) along with two artificial intelligence (AI) techniques, specifically Artificial Neural Networks (ANN) and Neuro-Fuzzy Inference System (ANFIS). A total of 10 daily metrological variables, for 13 years, namely Maximum temperature (Tmax), minimum temperature (Tmin), maximum humidity (Hmax), minimum humidity (Hmin), wind speed (Ws), wind direction (Wd), cloud cover (Cv), sea pressure (SEAp), station pressure (STAp) and relative humidity (RH) are used to predict precipitation (P). Both AI systems showed acceptable results predicting daily precipitation from observed meteorological parameters with a coefficient of determination (R²) of 0.75 and 0.72 for model calibration of ANN and ANFIS methods, respectively. The results of the ANN and ANFIS testing methods were 0.55 and 0.62, respectively. Outcomes of the study showed that the ANN model may have overfitted the results in the calibration section of the process compared to the ANFIS method, which performed better in the testing section of the evaluation process. In ANFIS modelling, for several input variables up to 6 variables, this study recommends using the grid partition method to divide variable ranges into membership functions. For input variables of more than 6 variables, sub-clustering method is recommended.