Mar. 6, 1995 to Mar. 10, 1995
R. Agrawal , IBM Almaden Res. Center, San Jose, CA, USA
R. Srikant , IBM Almaden Res. Center, San Jose, CA, USA
We are given a large database of customer transactions, where each transaction consists of customer-id, transaction time, and the items bought in the transaction. We introduce the problem of mining sequential patterns over such databases. We present three algorithms to solve this problem, and empirically evaluate their performance using synthetic data. Two of the proposed algorithms, AprioriSome and AprioriAll, have comparable performance, albeit AprioriSome performs a little better when the minimum number of customers that must support a sequential pattern is low. Scale-up experiments show that both AprioriSome and AprioriAll scale linearly with the number of customer transactions. They also have excellent scale-up properties with respect to the number of transactions per customer and the number of items in a transaction.
retail data processing; knowledge acquisition; pattern matching; very large databases; customer transactions; large database; customer-ID; transaction time; sequential pattern mining; algorithms; AprioriSome; AprioriAll; scale-up properties
R. Agrawal, R. Srikant, "Mining sequential patterns", ICDE, 1995, 2013 IEEE 29th International Conference on Data Engineering (ICDE), 2013 IEEE 29th International Conference on Data Engineering (ICDE) 1995, pp. 3, doi:10.1109/ICDE.1995.380415