2
$\begingroup$

I am currently trying to find out what algorithm IBM SPSS Modeler is using during calculation of decision lists. However, on the help page I can only find options for using the decision-list function but no information on how this model is trained. I also found some papers about decision lists, but these mention multiple different training methods. Is there any further information about the actual algorithms used by SPSS Modeler or are these completely kept as a secret?

NOTE:

I am not just interested in any algorithm for decision lists. Rather, we are currently basing our work on a decision list provided to used based on SPSS Modeler and we would like to start of using the same exact methodology as a starting point. For this it would be crucial to use the same algorithm that is implemented in SPSS Modeler.

$\endgroup$
2
$\begingroup$

You need the "Algorithms Guide" pdf file from the SPSS Modeler documentation. For Modeler 18.0, the decision list algorithm is on p.106.

The .pdf can be freely downloaded from IBM's website. Just google the documentation for your version of SPSS Modeler.

$\endgroup$

We're looking for long answers that provide some explanation and context. Don't just give a one-line answer; explain why your answer is right, ideally with citations. Answers that don't include explanations may be removed.

  • 2
    $\begingroup$ This is being automatically flagged as low quality, probably because it is so short. At present it is more of a comment than an answer by our standards. Can you expand on it? We can also turn it into a comment. $\endgroup$ – gung Feb 2 '17 at 12:19
  • 1
    $\begingroup$ This does not provide an answer to the question. Once you have sufficient reputation you will be able to comment on any post; instead, provide answers that don't require clarification from the asker. - From Review $\endgroup$ – mdewey Feb 2 '17 at 14:14
  • 1
    $\begingroup$ @mdewey I believe it does provide an answer. This question contains everything need to find the information to implement the algorithm myself. The spss documentation, that Borsislav points to gives a detailed description of the algorithm. The only better explanation would be to give some papers of where this algorithm comes from (not in the documentation), or include the actual algorithm. However, taking an algorithm from an IBM documentation may cause copyright issues. $\endgroup$ – LiKao Feb 10 '17 at 7:53
  • 1
    $\begingroup$ @LiKao: Is it worth preserving for posterity the information that an algorithm used by some software is described in the documentation for that software? We try to encourage answers that are going to be useful to others besides the one who asked the question. $\endgroup$ – Scortchi Feb 10 '17 at 9:49
  • 1
    $\begingroup$ LiKao: Thanks! Of course I won't presume to give you legal advice; but I look forward to your post. $\endgroup$ – Scortchi Feb 10 '17 at 10:35

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.