Process Documents from Data

Prentice

Hello,

I have a question regarding the "Process Documents from Data" and "Generate TF-IDF". 
What is the difference between the vector creating from "Process Documents from Data" set to TF-IDF and the "Generate TF-IDF". They give different end values, while they should both give the TF-IDF. So if I want to get the TF-IDF should I use the designated operator for this or the vector creation which is set to TF-IDF?

Thanks
-Prentice

MartinLiebig

Hi,

i wouild opt for Process Documents over Generate TF-IDF. the Generate TF-IDF is not normalizing the vector, thats why the results are different.

Best,

Martin

MartinLiebig

Hi,

i wouild opt for Process Documents over Generate TF-IDF. the Generate TF-IDF is not normalizing the vector, thats why the results are different.

Best,

Martin

Prentice

Ah ok thanks