The present work introduces the GADBMS, a system that uses a special restricted genetic algorithm to induce a set of rules from relational databases. The choice of GA paradigm is partially justified by its great capacity in dealing with noise, invalid or inexact data, and its easy adaptation to different domains of data. The GA algorithm uses Tabu lists to restrict the selection process. This restriction allows the creation of a set of potential rules for the classifier tool. This tool was tested in four datasets and compared to other twenty-three rules based algorithms. After that, noise was added to the databases and a new set of experiments was performed. The results prove that the proposed algorithm is efficient and robust. And the strategy used to maintain the diversity was considered valid, since the algorithm was able to keep its accuracy in categorization even for smaller populations.