Options

Text as a Data Type

dvvilkinsdvvilkins Member Posts: 1 Newbie
RapidMiner has a nice blog post on data types conversions (that RM won't let me link to as a noob) that classifies its data types. Only thing is that it doesn't mention text as a data type. However, RM tutorial process for the Nominal to Text operator clearly mentions text as its own as a data "type" and distinguishes it from nominal.  This leads to three questions:
  1. Where does text fit into the image below from the RM blog?
  2. How would I know if my data is text versus nominal/polynomial? Would I be able to see that from the statistics tab?
  3. Where does the concept of string values fit into all this? from the NTT operator description: "Also, the description for Nominal to Text operator says' This operator changes the type of selected nominal attributes to text. It also maps all values of these attributes to corresponding string values."

Sorry for all the questions but the RM documentation is lacking on this front.

  

Answers

  • Options
    MartinLiebigMartinLiebig Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert, University Professor Posts: 3,507 RM Data Scientist
    Hi there,

    here is the thought behind it:
    • Nominal types are types for different groups of items. For example every person needs to travel either in First, Second or Third class.
    • Text on the other hand are unique.Two news paper articles should never be an exact replica.
    Thats why text would be independent to nominal and why Text maybe handled differently than Nominal types. Also one may think about storing texts and nominals differently. But thats a bit of a different story and to the best of my knowledge not the case.

    Cheers,
    Martin
    - Sr. Director Data Solutions, Altair RapidMiner -
    Dortmund, Germany
Sign In or Register to comment.