Skip to Content
0
Former Member
Aug 02, 2016 at 08:21 PM

Match Transform - Driving Record Selection

114 Views

I have a data set as follows

id State Street City Full Name Dept 908 Nevada 0 Becker Avenue Reno Gregory West PLAN 183 Nebraska 0 Cascade Plaza Omaha Gregory Grant PLAN 538 Oklahoma 0 Dahle Pass Tulsa Shawn Lawrence PLAN 246 Illinois 0 Erie Alley Chicago Shawn Lawrence PLAN 9008 Nevada 0 Becker Avenue Reno Gregory West IT 1893 Nebraska 0 Cascade Plaza Omaha Phyllis Adsit IT 1893 Nebraska 0 Cascade Plaza Omaha Phyllis Adsit IT 5338 Oklahoma 0 Dahle Pass Oklahoma City Shan Larance IT

I want to match on Full Name, only those records with DEPT = "PLAN" with records with DEPT = "IT". In other words, records with DEPT = "PLAN" should NOT be matched with themselves and records with DEPT = "IT" should not be matched with themselves. Currently all the data resides in one table.

So in my result set will not have the following

  • Gregory West (in department = PLAN) should NOT generate a partial match with Gregory Grant
  • Shawn Lawrence from Tulsa should NOT generate a 100% match with Shawn Lawrence from Chicago.
  • Phyllis Adsit should NOT generate a 100% match with duplicate record Phyllis Adsit, both in IT department.

My result set should have the following

  • a 100% Match between record for Gregory West in PLAN department and IT department
  • a partial match for Gregory Grant & Gregory West in IT department.
  • a partial match for Shawn Lawrence and Shan Larance in PLAN and IT departments.

Any help in this matter is greatly appreciated.