Hi I am trying to use MS clustering algorithm to segment our customers, just based on customer purchase behavior, we have two tables:
--customer table
create table dm_msdm.scmolstp_city_mcc_accounts(
olst_txn_acct_nbr character varying(19) NOT NULL
);
--customer purchase table
create table dm_msdm.scmolstp_city_mcc_aggregated(
olst_txn_acct_nbr character varying(19) NOT NULL,
olst_mcc_code numeric(4,0),
olst_mcc_count integer,
olst_mcc_total_amt numeric(11,2));
I learned that the MS_Sequence_clustering algorithm can deal with nest table , but in my case the sequence of stores(olst_mcc_code) is not important, how can
I do this.
SQL Server 2008 GUI does not allow us to choose the MS_sequence_clustering algorithm, but only the clustering algorithm, does it mean if we use nest tables then
MS_sequence_clustering algorithm is automatically chosen?
View Complete Post