6 Replies Latest reply: May 15, 2014 10:08 AM by robeye _ RSS

    Using Monarch for fraud detection

    robeye _

      Hello,

       

      I have just started using Monarch so my abilities are quite limited.

       

      I have created a model that includes data from multiple accounts. What I want to attempt is to match/view only the transaction data common to the accounts included within the reports I import.

       

      Example: The reports contain transactional data for 3 accounts. I want to know what transaction(s) all three accounts have in common.

       

      Any help is greatly appreciated!:)

        • Using Monarch for fraud detection
          Olly Bond

          Hello Robeye,

           

          Welcome aboard! It would help a lot if you could post a sample of your data between and tags (without the spaces) so we could see what we're dealing with. Depending on the data, you might be able to do this with a filter, or you might need an external lookup.

           

          Best wishes,

           

          Olly

            • Using Monarch for fraud detection
              robeye _

              Advance Stores 3980 137th Ave Sw Miami Flus     88888

              Advance Stores 3980 137th Ave Sw Miami Flus     33333

              Albertso 8030 Pines Blvd Pembroke Pin Flus     33333[/CODE]

               

              In the above sample, I would like to try and filter or capture the common locations shared by the associated accounts. In this example, I would only want to see Advance Stores mathched with accounts 88888 and 33333 in the final summary along with any other matches that may be present within the rest of the report data.

                • Using Monarch for fraud detection
                  Olly Bond

                  Hello Rob,

                   

                  If you could negate a filter that was based on unique values, then a single pass would solve this. As it is, I think you're going to need two passes. First of all, create a summary using your two fields and the count() function, with matching on the second field set to measure => 2 to eliminate unique records.

                   

                  Then export this as an Access mdb table. Open your report again, make an external lookup, joining on the key fields, bringing in the count. Filter this to exclude non-null values, and you should have the right data set.

                   

                  HTH,

                   

                  Olly