Educational Data Mining is an area where in a combination of techniques such as data mining, machine Learning and statistics, is applied on educational data to get valuable information. The main objective is to recognize competent students based on marks are using clustering (X-means algorithm); then the subjects studied by them are classified into different categories and finally better combination of students as groups or teams are chosen to represent college for contests using association rules. To assess the performance of the proposed model, a student dataset of MCA from a college in Bangalore were collected for the study as a synthetic data. The accuracy of the results obtained from the proposed model was found to be promising. It was found from the study that 3 groups of 2 teams per group emerged as better combinations.