0
$\begingroup$

Let's say I have a primary dataset that its secondary dataset is hundreds to match and group like an one-to-many relationship.

I'm new in this world of the AI but my problem is that many child groups contain the same elements or even different combinations to result the parent data but in this case more to avoid duplication is get those duplications and the some way add up the data.

This is an example of what secondary data can look like and what I want to get from grouping it.

Parent data

  ID        FIELD1       FIELD2 FIELD3  FIELD4      FIELD5
  90148001  BLABLA       40     0       35896.89479 35896.89479

Child data

  ID        FIELD1       FIELD2 FIELD3  FIELD4      FIELD5
* 90148001  BLABLA       1      1770    1769.572665 1769.572665
* 90148001  DESCRIPTION2 1      13146   13146.45284 13146.45284
* 90148001  BLABLA       1      2176    2176.435074 2176.435074
* 90148001  BLABLA       1      2306    2305.716285 2305.716285
* 90148001  BLABLA       1      2531    2531.271196 2531.271196
* 90148001  BLABLA       1      1147    1146.803622 1146.803622
* 90148001  BLABLA       1      1991    1990.613246 1990.613246
* 90148001  BLABLA       1      3641    3641.394446 3641.394446
* 90148001  BLABLA       1      2471    2470.8253   2470.8253
* 90148001  BLABLA       1      2247    2246.984815 2246.984815
* 90148001  BLABLA       1      2471    2470.8253   2470.8253

Could a neural network be able to process, aggregate, and group those quantities?

$\endgroup$
  • $\begingroup$ Are you talking about clustering in general? Can you please explain more precisely how the neural network should "process, aggregate and group"? $\endgroup$ – nbro Sep 14 at 10:15
  • $\begingroup$ @nbro yes, I mean, something like the k-means algorithm, but instead of it being applied to image classification or processing, it can be used to sum and group the child data to result in the parent data, because that would be comparing the child data to match the parent data and add up and group them $\endgroup$ – victcdva Sep 14 at 14:23
  • $\begingroup$ Something is still not clear to me. Why don't you just mix the "child data" to get the "parent data"? Is your problem that many "child groups" contain the same elements so you want to avoid duplication? It's also not clear why you need clustering, at this point, because you use clustering to split data and not to group it. $\endgroup$ – nbro Sep 14 at 14:56
  • $\begingroup$ @nbro yeah, sorry if I'm not clear, I'm new in this world of the AI and so on but yes my problem is that many child groups contain the same elements or even different combinations to result the parent data but in this case more to avoid duplication is get those duplications and the some way add up the data. $\endgroup$ – victcdva Sep 14 at 15:08
  • $\begingroup$ Please, edit your post to include these details. I think it's important you provide an example of what the child data may look like and what you want to obtain by grouping it. $\endgroup$ – nbro Sep 14 at 15:10

Your Answer

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

Browse other questions tagged or ask your own question.