Hi @AhmetYavuz -
The String to Document node only works on data of type string, whereas you created a table using the PDF Parser, which produces documents already. (That is, if I’m understanding you correctly - let me know if I’m not.)
We actually just published a blog post a couple of weeks back that deals with pre-processing of OCR data, along with creating bigrams for additional analysis. I realize you are using PDFs as input as opposed to images, but a lot of the concepts are similar. As with most of our blog posts, it comes with a workflow too - check it out here:
https://www.knime.com/blog/an-experiment-in-ocr-error-correction-sharing-treasure-on-the-knime-hub