Mining Approach

Discussion in 'microsoft.public.sqlserver.datamining' started by anonymous_user, Jul 5, 2007.

  1. I have a scenario in mind and was wondering if anyone had any suggestions on an approach.



    Lets say I have a dataset where I have captured various attributes about blade servers in a rack. I have internal temperature, fan speed, disk reads, cpu temperature and failure events. Each of these are continuous variables except for the failure event, which has only two states, 'true' or 'false' which indicates whether a failure occurred at that point in time.



    The table looks like this:



    MachineID Timestamp Temperature FanSpeed Reads CPUTemp Failure




    At the time of failure, only the MachineID, Timestamp and Failure values will be populated in the table. The failures at this time are not categorized or discriminated from each other in any way although they will be in the future. We're looking to use a mining algorithm to determine which variable is the best predictor of failure, or combination of variables. What do you think is the right approach? How might this approach change once the failures do become categorized and differentiated?



    Thanks,

    Frank
     
    anonymous_user, Jul 5, 2007
    #1
    1. Advertisements

  2. anonymous_user

    Dejan Sarka Guest

    Frank,
    You can do couple of things. You can measure time between two failures (TBF)
    for each machine, and then use Regression Trees (Decision Trees with
    continuous variables) to try to predict the TBF using all other variables.
    You can also play with input variables and discretize some of them. Also,
    you discretize all input variables, and use a simple pivoting technique in
    T-SQL or Excel to analyze mean time between failures (MTBF) over input
    variables. When you will have failure categories, you can use Decision Trees
    again to try to predict failure category using other input variables. You
    can use TBF and failure category as inpit and predicatable variables.
     
    Dejan Sarka, Jul 6, 2007
    #2
    1. Advertisements

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.