An Algorithm of Mining Frequent Itemsets in Pervasive Computing

Based on DHP (direct hashing and pruning) algorithm, this paper presents a kind of transaction-marked DHP algorithm (TMDHP for short) to mining frequent itemsets in pervasive computing. Each element of the itemsets and the transaction´s ID will be stored together in the hash-table. Using this method just need to access database once and avoids producing a deal of candidate itemsets. The experiments showed that the performance of the algorithm is better than the conventional apriori algorithm and the DHP algorithm, and has a big advantage for application in pervasive computing.