There is no package called 'quanteda', How can I extract keyword/keywords RRS feed

  • Question

  • I need to extract particular keywords (e.g. bag_of_keywords <- c("apple", "  red apple", "I love apple.") from comment data, as well the user who made the comment. For example


    user                                                                    comment


    user1                                     tom has an apple tree full of big apples

    user2                         I am going to NC tomorrow

    user3                          I eat apple everyday

    how can I do it in AML?


    Tuesday, January 19, 2016 8:45 AM


All replies

  • Hi Bipul,

    Have you tried to import the R package (quanteda)? https://social.msdn.microsoft.com/Forums/en-US/2688bccc-2f38-4611-b466-f66edc0b43c5/using-bioconductor?forum=MachineLearning

    You might also need to bring in all the package dependencies if they are not already pre-installed in Azure ML. Use this code to check the latest list of packages that are preinstalled.

    data <- data.frame(installed.packages())
    # Select data.frame to be sent to the output Dataset port

    It looks like the package has a list of dependencies: https://cran.r-project.org/web/packages/quanteda/index.html


    • Marked as answer by Bipul Mohanto Tuesday, January 19, 2016 6:08 PM
    Tuesday, January 19, 2016 3:25 PM
  • Hello Jaya,

    Thank you very much for your genius comment. It took me several hours to understand the word "package dependencies", I found quanteda package requires two other packages, Rcpp and RcppArmadillo. So what I did, attached all these three packages in .zip and connect with my R script. But finally I found Azure ML is 3.1.0 when quanteda needs 3.2.2. Please help, how else I can find the keyword/words from my data. I am following my another question as well to clarify my purpose.


    I am really happy for your help.



    Tuesday, January 19, 2016 10:49 PM
  • Hi Bipul,

    Apologies for the inconvenience with the version of R pre-installed in Azure ML.

    Pls feel free to add in your feedback to our forum: https://feedback.azure.com/forums/257792-machine-learning

    This gives the development team a good idea of what the users need while using our product.


    • Marked as answer by Bipul Mohanto Wednesday, January 20, 2016 5:47 PM
    Wednesday, January 20, 2016 3:17 PM