Question about Sequential Covering Algo

I am training a rule based algo (PRISM or CN.2) with n classes (y_1,y_2,..,y_n). All rules in the training RuleSet are in DFN form, like IF t_1 OR t_2 OR ... t_m THEN y_i (terms) , where t_1 == lit_1 AND lit_2 AND ... lit_n (literals), and so on. When predicting in the test sample, based on the RuleSet, my question is: 1) IF is it possible to have an instance (example) classified with more than one class (y) (the answer I guess is YES, it is possible). IF it is possible, How to proceed? (Do a major vote for the different classes?). Could u guys please provide me some links describing this type of "problem"?