DuluthSemEval2021Task11
This repository contains the system written for SemEval 2021 Task 11: NLPContributionGraph. The code in the directory titled Phase1Code contains the end-to-end system solution. The folders Preprocessing and TrainingPipeline have already been run; their data is included in the DataFiles folder. To get the NLPContributionGraph datasets required to run the system on the SemEval contest test data, clone this repository. The testing pipeline relies on Stanford Core NLP for dependency parsing. To use, please download CoreNLP4.2.0 from here https://stanfordnlp.github.io/CoreNLP/index.html. Next, run ./download-data.sh and source launch-corenlp.sh &. To run the system on the test data, run Phase1Code/TestingPipeline/main.py. The submission data will be output to a folder called submission.