DBA Data[Home] [Help]

SYS.DBMS_DATA_MINING_TRANSFORM dependencies on STANDARD

Line 302: -- mean and standard deviation that are estimated from the data table:

298: -- DESCRIPTION
299: -- For every NUMBER column in the data table that is not in the exclusion
300: -- list finds normalization definition and inserts it into the definition
301: -- table. Definition for each relevant column is computed based on the
302: -- mean and standard deviation that are estimated from the data table:
303: -- shift = mean
304: -- scale = stddev
305: -- The values of shift and scale are rounded to round_num significant
306: -- digits prior to storing them in the definition table.

Line 599: -- standard deviation (dev)

595: -- definition table. Definition for each relevant column is computed using
596: -- equal-width method (see description for insert_bin_nume_eqwidth). The
597: -- number of bins (N) is computed for each column separately and is based
598: -- on the number of non-NULL values (cnt), min and max values, and the
599: -- standard deviation (dev)
600: -- N = floor(power(cnt, 1/3)*(max - min)/(C*dev))
601: -- where C = 3.49/0.9. Parameter bin_num is used to adjust N to be at
602: -- least bin_num. No adjustment is done when bin_num is NULL or zero.
603: -- Parameter max_bin_num is used to adjust N to be at most max_bin_num.