I'm trying to experiment if an opportunity will win or lose in Azure Machine Learning studio. However, am still in Data preparation method.

In my Data base I have opportunity table and products table.

For example, one opportunity has multiple products. Should I deal with the many products and put them in one record?

Will it affect the prediction if we have duplicate records for an opportunity like (a) or it’s better to have one record per opportunity in order to feed it to ML studio. And if yes which one will be better approach (b) or (c).

Approach a

enter image description here

Approach b

enter image description here

Approach c

oppid |first product |first technology|2nd product |2nd technology
1      out-services   active directory   TRN-items Adobe Acrobat 

The simplification of the data may make the model more stable, but it will also remove its ability to use more specific input criteria. For example, the way you move from Approach A to Approach B you are aggregating specific products into product categories. This means that if your model is successful in Approach A, it will be able to predict based on specific products. On the other hand, if your model training succeeds in Approach B it will only be able to predict on product category (and you will have to convert products into its categories before supplying it to the model).

So to answer your questions, the number of data samples you have determine how much you have to aggregate and simplify your data. The data itself in its most detailed form could also fail to train the model properly, in which case the approach taken in Approach B is the best next step.

  • $\begingroup$ Thanks @snympi , however, what do you think about approach c, is approach b better ? $\endgroup$ – sara May 14 '17 at 18:12
  • $\begingroup$ The difference between Approach B & C is not different from Azure ML's point of view (if I understand what you proposed with Approach C correctly). B and C is presentation of the same data in different formats - please correct me if not so. $\endgroup$ – Snympi May 15 '17 at 8:07
  • $\begingroup$ yes its the same data, however, in B its an aggregation of products ,technologies while in C its getting all data in one row for a particular id. $\endgroup$ – sara May 15 '17 at 11:01
  • $\begingroup$ I understand. In that case as far as Azure ML is concerned, Scenario B & C would yield the same results because the same data is essentially provided to the model - just in a different format. I think it will be more difficult or get the data into Azure ML because Scenario C's upload would be tricky because all the data is essentially one record and you will have to pull it apart as part of the data preparation. $\endgroup$ – Snympi May 15 '17 at 16:06
  • $\begingroup$ I understand that I should not follow approach C because its taking all data in one record right? . However, approach A in it the data is apart but as I understand Ill go with approach B where it count how many products per ID and put them in a category of numbers. Just to clear my mind, I thought that scenario A wont be convenient because if we have 10 product then we will have 10 rows, where in C it will be one row per id. But having the products apart is better like scenario A? $\endgroup$ – sara May 16 '17 at 6:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.