Although data mining represents the computational method of data processing, its use in education is still relatively new, i.e. its use is intended for discovering implicit, previously unknown, and useful knowledge out of existing data with an aim to make quality decisions in function of improvement of education system. The study was conducted by surveying the population of high school students in Tuzla Canton, Bosnia and Herzegovina (sample included about 10% of the student population, i.e. 1645 student). Using four different data mining algorithms the aim was to develop a model which can derive the conclusion of secondary level students' success.