Data Science Project for Information Extraction. Authors: Marcel Hoffman, Nicolas Lell, Steffen Epp, Michael Mohr.
This repository contains the code and additional artifacts for STEREO.
In the Code directory the python files for each step of STEREO can be found. The code files for GBCE is located in Code/GBCE and the ABEA files are located in Code/ABEA. The code files for the statistical extraction part are located directly in the Code directory.
For used verions please refer to the Requirements.txt. The installation manual can be found in the Technical_Report.pdf.
The original paper version of this work can be found here https://arxiv.org/abs/2103.14124 The paper was also accepted for the iiWAS2021
There is a follow up work (analyzing how much the set of R- rules can be reduced, how well STEREO works in the HCI domain, and PDF vs LaTeX as input), see https://arxiv.org/abs/2211.13632 or https://github.com/Tobi2K/BachelorThesis