December 21, 2009 at 7:40 pm
Hi,
I have 802K records in a dataset and a reference table. I have two fields: schoolcode and schoolname. I have a reference table with schoolcode and schoolname in it with some other data I need in my original table.
The trouble is, they just don't match up exactly in all cases. Some of the differences are predictable while others aren't.
I ran fuzzy lookup and sent the results to another table to look at how they match up.
There doesn't appear to be a similarity threshold that would automatically just get me the dataset with the automatic matches.
I created a view with all of the distinct code and name matches. I'd like to add a column such as 'isAccpeted' or something like that. This way, I can look over each distinct match and put a 1 for the ones I want and a 0 for the one's I don't.
What I'm not sure about, however, is how to send those changes BACK to the original table. Because my intermediate view only shows distinct values, I would only be marking each match, not each row.
What kind of thing would I need to do to send those changes back to the original table? I've thought of perhaps using the Merge task or maybe the Join task. Any ideas?
Thanks,
Ron
March 1, 2010 at 11:06 pm
I worked on some of the large scale fuzzy matching projects. Based on my experience you are in the right track.
You already have a table to which you load your fuzzy results. Add a derived column named IsMatchFlg and default the value to 0. Once you finish loading the data into the table run a SQL script to flag you aceptable record as IsMatchFlg=1.
This flag should be applied to 1 record per batch which is the record having maximum similarity score.
Then a select from this table where IsMatchFlg=1 will give your matching records.
I guess this helped. Let me know
Viewing 2 posts - 1 through 1 (of 1 total)
You must be logged in to reply to this topic. Login to reply