"how to use knn,svm in real time"

magedtab
magedtab New Altair Community Member
edited November 5 in Community Q&A
Hello RapidMiner-Community,

Can i use meta-learning algorithms like knn ,svm in real time if it is possible , what's steps?
Now, I would be very thankful if you could give me some technical advice .

Thanks for your help!

Greetings,
Tagged:

Answers

  • magedtab
    magedtab New Altair Community Member


    hi every one, i need your help in my project and is to implement stacking operators in real time  ,
    the problem is i don't know how to start ,so i need your ideas and support to do this.i collected data and used rapidminer in offline mode .now i need to do the same work but in real time by using stacking .


    i am waiting your help

    majed
  • MariusHelf
    MariusHelf New Altair Community Member
    Hi majed,

    can you describe your problem a bit more detailed?

    What do you mean by stacking? How do you define real time?

    Best regards,
    Marius
  • magedtab
    magedtab New Altair Community Member
    Dear friends
    thank you Marius
    i did classification by using stacking operator in offline mode .
    now i want to make the same but in real time .
    real time which controls an environment by receiving data, processing them, and returning the results sufficiently quickly to affect the environment at that time
    my project detect DOS attacks in real time by using data mining . (online )
    image link show (process of stacking done by rapidminer).
    image
    keep in touch

    thank you friends
  • MariusHelf
    MariusHelf New Altair Community Member
    Hi,

    the X-Validation has a model output port. Store the resulting model via the Store operator to the repository.

    Then create a new process. In the new process you can load the stored model using the Retrieve operator, and apply it to new data in the same way as in your testing process, i.e. with the Apply Model operator.

    That helps you to apply the model in real time. Does that help?

    Best regards,
    Marius