hi,
i have the following scenario.
i have an incoming source file which has multiple records which r not duplicates but consist of transaction data for the same person. And i have the existing file which has records of unique persons.
i need to match both the files. which MATCH strategy do i need to use.
match sets or geomatch multiple????
and when i do that which shud b file A and which shud b File B
od give me some idea
thanks in advance.
which match strategy do i need to use
-
- Participant
- Posts: 3593
- Joined: Thu Jan 23, 2003 5:25 pm
- Location: Australia, Melbourne
- Contact:
The transaction file should be File A and the person file should be File B. Your objective I assume is to read in transactions and attach to them additional personnel details from the person file.
What is your matching field? Is it a persons name or is it some type of ID field? How unique is your "unique" person file. For example how do you uniquelly identify "John Smith" when there are several people who may have that name.
What is your matching field? Is it a persons name or is it some type of ID field? How unique is your "unique" person file. For example how do you uniquelly identify "John Smith" when there are several people who may have that name.
Certus Solutions
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
Blog: Tooling Around in the InfoSphere
Twitter: @vmcburney
LinkedIn:Vincent McBurney LinkedIn
which match strategy do i need to use
hi, thanks for ur reply.
actually i have 2 files. the matching fields in the file is the Date_of_birth and SSN(number). the file A (say the transaction file) has say for example 10 records belonging to a person. But what i thought, e sacond file say File B which i said was unique is nt so. like for a single person there are may b more than one record in which all the columns consist of the same information excepting one column which is RXID.
so the file B is also not unique.
i eed to match the two files and append the concerned columns to the file existing
the transactions file has 90000 recs and the file b has 1900 recs.
thanks!
actually i have 2 files. the matching fields in the file is the Date_of_birth and SSN(number). the file A (say the transaction file) has say for example 10 records belonging to a person. But what i thought, e sacond file say File B which i said was unique is nt so. like for a single person there are may b more than one record in which all the columns consist of the same information excepting one column which is RXID.
so the file B is also not unique.
i eed to match the two files and append the concerned columns to the file existing
the transactions file has 90000 recs and the file b has 1900 recs.
thanks!
g.kiran
Re: which match strategy do i need to use
Do you need the RXID column of file b? Or a particular record associated with a particular RXID (e.g. the max RXID)? Dependingly, it would be worth cleaning up file b first to be unique records.DSkkk wrote: there are may b more than one record in which all the columns consist of the same information excepting one column which is RXID.
so the file B is also not unique.
i eed to match the two files and append the concerned columns to the file existing
Geomatch mutiple and geomatch duplicate will allow linking to more than one file b record, the difference being in geomatch multiple that duplicates only count if they have the same weight. (If RXID isn't used, then they should). You can then use geomatch multiple to link to all file b records. Remember, though, that the duplicate file b records will be in the DUPB extract file, not in the MATCH extract file.
(Geomatch plain will just link to one file b record and ignore any duplicates)
(I used geomatch duplicate to let me know about my false positives in one case and because they were valid links in another.)
Cheers,
Jamas
which match strategy do i need to use
thanks James,
i have got the olution.i have used the geomatch multiple strategy.
thank u
i have got the olution.i have used the geomatch multiple strategy.
thank u
g.kiran