The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here

"Sequence and error analysis"

JCJC Member Posts: 2 Contributor I
edited May 2019 in Help
Hi,

I am new to data mining and I am wondering if someone can point me in the right direction. I apologize for the long explanation, I just want to make sure that I am clear.

I am using RapidMiner to analyze what users did while using an educational program. I had 90 users, each user completed 20 exercises. The system created a log of the time of every click, also classifying each action as a command, help item, resource or error, for each exercise. Since every user clicks on different items, the number of records (rows) is not the same for each user, it ranges between 170 and 600. I do however have the same number of attributes for all of them.  I know it is similar to web usage, but in this case I don't have only pages. I want to compare the sequences of actions between users, how they used the available resources, and the errors they made (basically I am comparing the click sequences). I want to make this comparison for the entire session, but I would also like to compare each exercise individually accross users. Is there a technique or set of techniques that can help me in this analysis?

If I cannot achieve this with RapidMiner but you know another technique or software in any domain that could be used, I would greatly appreciate any suggestions.

Thanks for your help!
Tagged:

Answers

  • landland RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 2,531 Unicorn
    Hi,
    seems to be an interesting application for data mining. A nice variety of the usual financial stuff. I'm quite sure you can do something with RapidMiner, since I know, that it has been used for WebUsage mining in the past and the problem seems to be quite equal, although having different attributes. But since I never have worked on such a problem myself, I can't give you any qualified hint, beside from pointing on the Text Plugin, which contains some operators for translating logs to usable example sets. (LogFileSource and ServerLog2Transactions)
    And all in all the topic seems to be too complex for answering in the forum, although your explanation is exemplary :) If you need real case based help, you could ask for telephone consulting, which is often the ideal form for such problems, because you not only get pointed in the right direction, but in fact you receive some hand crafted processes for your special problem.

    Greetings,
      Sebastian
  • JCJC Member Posts: 2 Contributor I
    Thanks for your advice, Sebastian! I will check the Text Plugin and see if that does the trick, and if that does not work, then I will go for the telephone consulting session. Since it has been a bit difficult for me to play with RapidMiner, I will most likely post video tutorials on my site when I have found out a way of doing the analysis. It might help others who are not working in finance.
Sign In or Register to comment.