Repository navigation
[data] Collect and compile glossaries available from PDFs and other sources used in WASH and related fields #54
Description
This idea builds on process of the R data package glossarywho, which was published from a PDF document shared by the World Health Organisation.
Glossaries are being written for almost every report that is prepared in the WASH sector and related fields. Rarely, existing definitions are used, but terms are redefined by the authors, often without a reference to the original source.
This issue is an idea to:
- Collect available WASH resources (reports) that contain glossaries
- Extract terms and definitons from resources and compile them into tidy dataframes, following the openwashdata workflow
- Identify misaligned definitions and irregularities for the same usage of terms
- Discuss whether alignment is of terms would be beneficial or whether terms in single reports can work within that context
Activity
larnsce commentedon Feb 19, 2025
larnsce commentedon Feb 19, 2025
Pahulmeet commentedon Mar 17, 2025
Proposed Solution for Glossary Extraction
Code: https://github.com/Pahulmeet/extract_glossary
Hello,
I would like to propose the above solution to extract and structure glossary data from PDFs. I believe this approach can address the challenges of misaligned definitions and irregular formatting while providing results in seconds.
More details about the complete approach and solution can be found on a README.md file in my code directory on GitHub named extract_glossary.
Note: I hope my comment does not affect any official workflow of the openwashdata community guidelines. The issue was open and unassigned so giving it a shot :)