Attribute-wise Learning (ALSO)

GottfriedGottfried Member Posts: 17 Maven
Hello Community,
I have a hard time trying to get outlier scores from the ALSO operator. Whatever my tweaking, the dataset, the internal regression model… the result is a useless long list of zero outlier scores.
Thanks in advance for your help!

Answers

  • MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,485 RM Data Scientist
    Hi Gottfried,
    from which extension does this come from? Honestly - it is the first time i hear about it. Good to learn something new!

    BR,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
  • GottfriedGottfried Member Posts: 17 Maven
    The extension is called "Anomaly Detection". It contains 17 operators. It notably offers a very efficient "Replicator Neural Network" which is today the closest thing to the implementation of an autoencoder you can get within RapidMiner. The RNN actually runs a neural autoencoder to score outliers but does  unfortunately not give access to the autocoding itself (the dimensionality reduction). But this is another story. The ALSO operator on the other hand seems bugged… Too bad!
  • sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager
    hi @Gottfried yes that's an AGPL extension so it's subject to the author(s). You can maybe ping them on GitHub and see where that goes? https://github.com/Markus-Go/rapidminer-anomalydetection

    Scott

  • GottfriedGottfried Member Posts: 17 Maven
    I followed your advice… Let’s see.
    Gottfried
Sign In or Register to comment.