Practical Problem Solving - Java Data Mining: Strategy, Standard, and Practice - page 318

5. throws SQLException {

6. mCustomerTableName iCustomerTableName;

7. mColumnNames iColumnNames;

8. mIdentifierColumnName iIdentifierColumnName;

9. mOutputTableName iOutputTableName;

10. String lSQLQuery "create table " mOutputTableName " as select ";

11. lSQLQuery mIdentifierColumnName;

12. for (Iterator lColNamesIter mColumnNames.iterator();

13. lColNamesIter.hasNext();) {

14. String lColName (String) lColNamesIter.next();

15. lSQLQuery ", " lColName;

16. }

17. lSQLQuery " from " mCustomerTableName;

18. Statement lStatement mJDBCConnection.createStatement();

19. lStatement.executeQuery(lSQLQuery);

20. String lSQLCountQuery " select count(*) from " mOutputTableName;

21. Statement lStatementCount mJDBCConnection.createStatement();

22. ResultSet lResultSetCount lStatementCount.

23. executeQuery(lSQLCountQuery);

24. lResultSetCount.next();

25. int lSamplingCount lResultSetCount.getInt(1);

26. mCurrentState CustomerSegmenterState.EXPORTED_FOR_ENHANCEMENT;

27. report("Count for enhancement: " lSamplingCount);

28. }

This code generates the SQL statement that creates the desired
table from the select statement as shown in lines 10 to 17. The number
of records produced is reported to the user, because it may be linked
with payment to TSB.
Then some time is needed by TSB to enhance the data and return
the enhanced file, which is loaded by the IT team into the OMDB.
This enhanced table is used as the basis for building the clustering
models. As described for this scenario, several clustering models
with different numbers of clusters are built. This is done by the
method called buildModelAndApply . Once the clustering models are
built, they are used to assign each customer of the enhanced table
with their cluster identifiers.
In the buildModelAndApply method, the parameter iModelPrefix is
used together with the number of clusters found in that model to
generate the model names. For example, if the prefix is “MyModel”
and iMinClusterCount is set to 5, and iMaxClusterCount is set to 7,
then the models “MyModel5,” “MyModel6,” and “MyModel7” will
be built by this routine. The argument iEnhancedTableName is the
table name containing demographic and psychographic data filled
by the IT department from the file returned by TSB.