1
$\begingroup$

I have a requirement where I need build a solution to extract information from multiple pdfs. The information is stored in tabular format in pdf files. The tables are not very well structured and each pdf document has different format of tables. The pdfs consist of a lot of pages with text and tables.

I tried Tabula, but it didn't produce a satisfactory result. Thought about using pdfminer as well, but it looks very cumbersome to me.

Can you suggest more ideas and tools to solve this problem?

$\endgroup$
2
  • 2
    $\begingroup$ what platform are you on R, Python...? $\endgroup$ – Peter May 31 '19 at 11:35
  • $\begingroup$ try algodocs.com $\endgroup$ – mussdroid Aug 1 '19 at 0:15
0
$\begingroup$

For some literature on the problem you may want to look at https://medium.com/sap-machine-learning-research/chargrid-77aa75e6d605 and https://arxiv.org/abs/1809.08799

For a commercially available solution Textract and Tika come to mind.

$\endgroup$
0
$\begingroup$

Pdfminer, Pypdf2, Camelot, Tika, OCR, Tesseract, Textract, Nuance OCR

These are a few pdf parsing tools.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.