InaPlex Forums

Hi I am using Inaport 7 with GoldMine and need to use Fuzzy Matching I would like to be able to match on Company using a Fuzzy match and contact using an exact match. However, I can't see that selecting the Fuzzy Matching option allows me to match multiple fields in the same way that the standard matching does. Thanks

Hi

You are able to set up match expressions in fuzzy matching in exactly the same way as you can in standard matching. So for example, you could set up a match like this:

norm(#srcComp & #srcCont)    <->    norm(#company & #contact)

This combines two fields together and normalies the result (norm() will lower case and trim strings)

A problem with GoldMine is that company and contact are in the same table, so it is difficult to set up different match criteria unless you put the "_contact1" table into the mapping tree twice - this is not recommended.

HTH. If not, please expand your question.

Regards

Admin

Hi Thanks for the reply, just to elaborate and check your response My source file will have multiple rows for the same organisation but with different contacts. Therefore only one of these rows is a correct match for the organisation and contact combined. The import needs to use fuzzy matching on the company since it is not always an exact match. However, I am happy to use exact match on the contact. So "Joe Scott of ABC Company" in the source will match with "Joe Scott of ABC Company Limited" in GoldMine but another row, "Jenny Black of ABC Company" will not match with the "Joe Scott of ABC Company Limited" will not match with GoldMine. So would what you've suggested work for this scenario? Thanks The objective is to assign an ID against the contact record.

Hi

As noted, because both company and contact are in the same table (in fact, same record) for GoldMine, it is difficult to set up different match criteria.

I assume that you will have records in contact1 for ABC/Joe and ABC/Jenny. What I have suggested will work for this.

If the only problems with company name are things like Inc, Corp, LLC, etc you might want to have a look at the normcomp() function, instead of fuzzy matching, normcomp() will remove punctuation and noise works such as Inc, Incorporated, ... as well as lower casing and trimming whiste space.

normcomp(#srcCom) & norm(#srcName)     <->    normcomp(#company) & norm(#contact)

Regards

Admin