Rule-based Automatic Question Generation Using Semantic Role Labeling

Code for IEICE 2019 paper Rule-based Automatic Question Generation Using Semantic Role Labeling

The designed system is written in Python 3.6. Parsers:

• -SENNA's part-of-speech tagging and chunking,
• -SpaCy's dependency parsing and NER,
• -AllenNLP's semantic role labeling

To install:

• python3 setup.py install
• pip3 install spacy==2.0.18
• python3 -m spacy download en
• pip3 install allennlp==0.6.0
• pip3 install scikit-learn==0.22.2

To run:

• cd pntl
• python3 tools.py

Edit _main_ in tools.py and add your own sentences to get corresponding questions