word list

hasnoooooor
hasnoooooor New Altair Community Member
edited November 5 in Community Q&A
i have a model that has been trained and tested with dataset a. now i am inserting a new dataset into the model and try to get the wordlist output. the problem is if i did not connected the wordlist output from dataset a, the missing attribute problem will appear. but, when i connected the old wordlist data along with the new datasets, i got the wordlist for dataset a and not for the new dataset.. how can i do if i want the wordlist for the new example set? 
Tagged:

Answers

  • MartinLiebig
    MartinLiebig
    Altair Employee
    Hi,

    did you read up on this? https://community.rapidminer.com/discussion/50278/text-mining-and-the-word-list 

    BR,
    Martin 
  • hasnoooooor
    hasnoooooor New Altair Community Member
    Hi,
    yes.. i already read it. currently i am doing this, and i would like to have an output of word list from the new input example set, but the result keep giving me the word list that i have for training.
  • kayman
    kayman New Altair Community Member
    The word list is to be seen as a filter, or in other words, new words are ignored as they were not included in the training and are therefore not relevant for the model. 

    Since your new data will be labeled using the words found during training only, it is therfore kind of logical the same words are returned. If you want to include new words it requires retraining of your data. 
  • Telcontar120
    Telcontar120 New Altair Community Member
    You could also convert your wordlist to data from the Toolbox extension and then use the normal Join operators with another dataset that you create that has only the words you want in it.

Categories