Fast Mining of Maximum Frequent Itemsets in Distributed Multimedia Database

Traditional algorithms for mining maximum frequent itemsets in local database and for mining global frequent itemsets do not suit mining of maximum frequent itemsets in distributed multimedia database. As far as we know, a little research of mining global maximum frequent itemsets has been done. Therefore, the paper proposes an algorithm for fast mining maximum frequent itemsets in distributed multimedia database, namely, FMMFI algorithm, which sets center node. FMMFI algorithm makes computer nodes compute local maximum frequent itemsets independently with DMFIA algorithm, then the center node exchange data with other computer nodes and combined, finally, global maximum frequent itemsets are gained. FMMFI require far less communication traffic by the search strategy of top-down. Theoretical analysis and experimental results suggest that FMMFI algorithm is fast and effective.