Data_Discovery_for_the_SDGs

Source code and data for the paper "Data Discovery for the SDGs: A Systematic Rule-based Approach".

Setup

You will need to install Poppler and Tesseract before you can run the Python code.

Python requirements can be found in requirements.txt.

Usage

There are two Jupyter notebooks:

• notebooks/01-Entity_Extraction.ipynb : Loads PDFs from data/sdg7-papers (note: PDFs note included in this repository), converts each to images using Poppler, then uses Tesseract to perform OCR to extract the text, then extracts entities based on rules derived from data/sdg7-coding-manual.xlsx to output sdg7-coding-auto.xlsx.
• notebooks/02-Plot_Figures.ipynb : Loads data/sdg7-coding-auto.xlsx and applies some processing before plotting sunburst chart of data mapping distributions.