Attribution_MLE Example (Single Input) | Teradata Vantage - Attribution_MLE Example (Single Input): Window Models - Teradata Vantage

Machine Learning Engine Analytic Function Reference

Product
Teradata Vantage
Release Number
9.02
9.01
2.0
1.3
Published
February 2022
Language
English (United States)
Last Update
2022-02-10
dita:mapPath
rnn1580259159235.ditamap
dita:ditavalPath
ybt1582220416951.ditaval
dita:id
B700-4003
lifecycle
previous
Product Category
Teradata Vantageā„¢

This example uses the single-input Attribution_MLE function.

SQL Call

SELECT * FROM Attribution_MLE (
  ON attribution_sample_table PARTITION BY user_id
    ORDER BY time_stamp
  USING
  EventColumn ('event')
  ConversionEvents ('socialnetwork', 'paidsearch')
  ExcludeEvents('email')
  OptionalEvents ('organicsearch', 'direct', 'referral')
  TimeColumn ('time_stamp')
  WindowSize ('rows:10&seconds:20')
  FirstModel ('SEGMENT_ROWS', '3:0.5:EXPONENTIAL:0.5,ROW',
          '4:0.3:WEIGHTED:0.4,0.3,0.2,0.1', '3:0.2:FIRST_CLICK:NA')
  SecondModel ('SEGMENT_SECONDS', '6:0.5:UNIFORM:NA', '8:0.3:LAST_CLICK:NA',          '6:0.2:FIRST_CLICK:NA')
) AS dt ORDER BY 1, 3;

Output

 user_id event         time_stamp                 attribution         time_to_conversion 
 ------- ------------- -------------------------- ------------------- ------------------ 
       1 impression    2001-09-27 23:00:01.000000 0.20000000298023224              -19.0
       1 impression    2001-09-27 23:00:03.000000                 0.0               NULL
       1 impression    2001-09-27 23:00:05.000000                 0.0               NULL
       1 impression    2001-09-27 23:00:07.000000                 0.0               NULL
       1 impression    2001-09-27 23:00:09.000000                 0.0               NULL
       1 impression    2001-09-27 23:00:11.000000                 0.0               NULL
       1 impression    2001-09-27 23:00:13.000000 0.30000001192092896               -7.0
       1 impression    2001-09-27 23:00:17.000000                0.25               -3.0
       1 impression    2001-09-27 23:00:19.000000                0.25               -1.0
       1 socialnetwork 2001-09-27 23:00:20.000000                NULL               NULL
       1 direct        2001-09-27 23:00:21.000000                 0.5               -2.0
       1 referral      2001-09-27 23:00:22.000000                 0.5               -1.0
       1 paidsearch    2001-09-27 23:00:23.000000                NULL               NULL
       2 impression    2001-09-27 23:00:29.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:31.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:33.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:36.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:38.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:43.000000 0.20000000298023224              -16.0
       2 impression    2001-09-27 23:00:47.000000                 0.0               NULL
       2 impression    2001-09-27 23:00:51.000000 0.30000001192092896               -8.0
       2 impression    2001-09-27 23:00:53.000000                0.25               -6.0
       2 impression    2001-09-27 23:00:55.000000                0.25               -4.0
       2 socialnetwork 2001-09-27 23:00:59.000000                NULL               NULL

Download a zip file of all examples and a SQL script file that creates their input tables.