Mining frequent closed itemsets provides complete and condensed information for non-redundant association rules generation. Extensive studies have been done on mining frequent closed itemsets, but they are mainly intended for traditional transaction databases and thus do not take data stream characteristics into consideration. In this paper, we propose a novel approach for mining closed frequent itemsets over data streams. It computes and maintains closed itemsets online and incrementally, and can output the current closed frequent itemsets in real time based on users' specified thresholds. Experimental results show that our proposed method is both time and space efficient, has good scalability as the number of transactions processed increases and adapts very rapidly to the change in data streams. Categories and Subject Descriptors H.2.8 [Database Management]: Database Applications ? Data Mining. General Terms Algorithms, Performance, Experimentation. Keywords Data stream, freque...