SQL & PL/SQL

Hi,

I would like to optimize the performance of the below query,

SELECT TO_NUMBER(TO_CHAR(date_col, 'iyyy')) AS paid_in_year,

                               TO_NUMBER(TO_CHAR(date_col, 'ww')) AS paid_in_week,

                               EXTRACT(DAY FROM ((date_col - date_col2) DAY(4) TO SECOND)) AS input_to_paid_days,

                               COUNT(id) AS ID,

   date_col2 as PAID_DATE

                          FROM big_table

                         WHERE extract(year from date_col2) >= extract(year from sysdate) - 2

                           AND col1 != 'NWLK'

                           AND col2 = '02'

                         GROUP BY TO_NUMBER(TO_CHAR(date_col, 'iyyy')),

                               TO_NUMBER(TO_CHAR(date_col, 'ww')),

                               EXTRACT(DAY FROM ((date_col - date_col2) DAY(4) TO SECOND)),date_col2

The table referred in the query is a very huge table. Please help me in fixing this.

I understand that it has got more oracle functions in it and and a group by that makes the query to run slower.

But please assist in restructuring the SQL. This is my base query and I have even more logical calculations run over it to get the final result.

Thanks.