pksm3 / pdfparsing

pour la recherche, pour la santé, pour demain

Usage:

(1) python extracter.py "Grant proposals"/

This will generate a folder, e.g. "TO-USE/", with lots of txts from GrantProposals' pdfs. Errors log in FailFolders.txt

(2) python processOutput.py

• Install http://github.com/saffsd/langid.py
• pip install openpyxl

This will read the txts folder, e.g. "TO-USE/", and it will generate a CSV with:

• ID: BPM-ID
• PI: P.I.-fullname defined by the merge/unification of the names that appear in GrantProposalsFolder and extractionBPM.xlsx
• AG1: Agency according GrantProposalsFolder
• AG2: Agency according extractionBPM.xlsx
• ABS: Artifitial abstract using the pdfs extracted-info (for some filtering: blacklist.txt)
• ACR: Acronym according extractionBPM.xlsx
• TI: Title according extractionBPM.xlsx

This CSV is uploaded to Cortext.