I work in a telco and manage the data in the company's CRM. I'm often given lists of company names by our alliance team, and asked to match the list against our database to see which companies already exist in the CRM and which don't. At the moment I can only accomplish this by one-to-one matching, using a LIKE "%part of company name%" search in Access, examining the results, and choosing the closest match. The lists given to me by Alliance often have hand type names, using ampersands, excluding the INC off the end, incorrect spelling, inappropriate punctuation etc - which makes a SELECT WHERE 'company name' is equal query completely fruitless. My question is this - can anyone recommend string matching software or an SQL query that will wash one list against another, finding the closest match with confidence intervals? I don't know if this would involve fuzzy logic or not - but I'm convinced there needs to be a more efficient way to go about doing this. Any advice is greatly appreciated, thanks a lot for reading.
April 1, 2009 10:54 PM
April 3, 2009 12:23 AM