What is data file format for operator W-GeneralizedSequentialPatterns

windpastwindpast Member Posts: 2 Contributor I
edited November 2018 in Help
I want to use operator Weka-GeneralizedSequentialPatterns, however the system always paused with the error message:
Error in: W-GeneralizedSequentialPatterns(W-GeneralizedSequentialPatterns)
W-GeneralizedSequentialPatterns caused an error:
Weka.associations.GeneralizedSequentialPatterns: Cannot handle missing values!

However I have no idea about the data file format that Weka-GeneralizedSequentialPatterns need.
I tried pure text file, csv file, xls file and spss file, everything caused the same error messge.
Even I delete all the contents of the file, and only leave one word such as "abcd", the same error message appears.

Please help me if you know how to solve the problem.
Thanks in advance.

Answers

  • haddockhaddock Member Posts: 849 Maven
    Hi Windpast,

    Being also a newbie I hesitate to offer an opinion, but hey it is Friday, and I'm not so bashful! As far as I see it you are asking the wrong question. All learners take as input an example set, and example sets are produced by I/O operators. So it doesn't matter what the file format is. As long as an I/O operator exists for the file format you can transform that file into an example set.

    I've not used the W-GeneralizedSequentialPatterns operator but it obviously does not like missing values in an example set, hence the error. If you use an I/O operator to make your example set, and then see question marks in some of the examples, you will get this error when you apply this learner. There are operators to filter out such examples, and other operators which replace the missing values. Full details in the glossy manual.

    Personally I found working through the tutorial very useful, it just gets you into the right mindset.

    Great weekend to all, time for a drink!
  • windpastwindpast Member Posts: 2 Contributor I
    Hi, haddock
    Thank you very much for your quick reply.

    Maybe I did not describe my question clearly enough.
    I did try some I/O operators (such as SPSSexampleSource and SimpleExampleSource) to read in my data file.
    Exactly, I used some other learner operator(e.g. w-Apriori) , and got some corresponding results successfully.
    However when I replace the learner operator with the W-GeneralizedSequentialPatterns, it did not work.
    I guess there must be something wrong with the data file format or usage of the operator,
    but I can not find enough information from the document and Tutorial to solve the problem.
    Maybe the last choice is to run RapidMiner in JAVA environment and try to trace into the JAVA source code of the learner operator?  Since I am not very familiar to JAVA, even I found the source code of the W-GeneralizedSequentialPatterns, I am not sure whether I can locate the reason of the error.
    Anyway, thank you again for your help.
Sign In or Register to comment.