Relationship between the pollution status and the distribution of aquatic fauna has been one of the most important issues in ecological research. Since this relationship involves rather complex nonlinear relations there are lots of novel modelling approaches to improve the performance of predictions. Artificial interligence and machine learning techniques played a crucial role in studying this type of relationships. Many techniques had been applied for macroinvertebrate modelling including decision trees, artificial neural network, fuzzy logic, genetic algorithm etc. Meanwhile, support vector machines have been successfully applied to solve nonlinear regression and time series problems. This investigation elucidates the feasibility of the support vector machines, to correlate the river water quality with the macroinvertebrate community composition. In this study, the macroinvertebrate distribution in the Du river in the North Vietnam is used to illustrate the convenience of support vector machines in this context.