Image Mining

shamini
shamini New Altair Community Member
edited November 2024 in Community Q&A
sir,
i'm doing image mining using rapidminer .
to do that, i've converted my images into grayscale , extracted text from images using ocr technique n now performing text mining to the text extracted .

while i perform text mining with tokenise operator, a stray character is always introduced n hence destroys the tokenising pattern i.e instead of tokenising it into words it tokenises it into letters 

how can i remove stray character
please help
from past 3 weeks , its bugging me ...any help or suggestions 
thanks in advance
Tagged:

Answers

  • alsaqer002
    alsaqer002 New Altair Community Member
    Hello Shamini,

    Try to use "Generate n-Grams (Terms)" in order to get words instead of letters.

    Would you please tell me how did you extracted text from images using ocr technique?

    Thanks,