This system adopts Artificial Intelligent (AI) approaches (of hybrid Kohonen Self Organizing Maps and Fuzzy Logic) in modelling and predicting the water quality in terms of the growth of coliform and e-coli for tropical lakes and natural wetlands of Putrajaya, Malaysia. The main motivation behind the invention is due to fact that the presence of bacteria in water particularly e-coli is hazardous and can cause serious deceases such as cholera, typhoid fever, dysentery, etc. Whereas on the other hand the presence of bacteria cannot be easily detected by human senses- thus, it is highly desirable to predict the water quality by easily observed ecological factors such as temperature, pH and biochemical oxygen demand (BOD) that is known to have significant influence to the growth of bacteria coliform in water. The overall system has been trained and tested for its accuracy on a reliable database that provides samples of relevant parameters from tropical Putrajaya Lakes and Wetlands over a period of five years. The system has demonstrated prediction up to 90.5% of accuracy.