The discovery of generalized fuzzy association rules is a very important data-mining task, because more general and qualitative knowledge can be uncovered for decision making. In the literature, few algorithms have been proposed for such a problem, moreover, the efficiency of these algorithms needs to be improved to handle real-world large datasets. In this paper, we present an efficient method named cluster-based fuzzy association rule (CBFAR). The CBFAR method creates cluster-based fuzzy-sets tables by scanning the database once, and then clustering the transaction records to the k-th cluster table, where the length of a record is k. Based on the information stored in the table, less contrast and database scans are required to generate large itemsets. Experimental results show that CBFAR outperforms a known Apriori-based fuzzy association rules mining algorithm.