Abstract:

To solve the problem that a large number of candidate sets will be generated when an apriori algorithm is used to scan the transaction database many times to look for frequent itemsets, a frequent itemsets mining algorithm is presented based on the combination of vector and array, which can scan the transaction database only once, avoid pattern matching and reduce the generation of worthless candidate sets. In addition, by comparison with the existing algorithms, this algorithm is verified with a high efficiency of mining. And the more items in the database the more effective it is.

Key words: data mining; association rules; apriori algorithm; frequent itemsets