We present an approach for automated hypothesis generation that can help the analyst explore the space of possibilities.

• Set up the Stanford's core nlp server instance on localhost:9000
• Follow instructions here: https://www.khalidalnajjar.com/setup-use-stanford-corenlp-server-python/
• Install python 3 and the package "stanfordcorenlp" using pip
• Install flask

• Set up the Stanford's core nlp server
• Open jupyter and run each cell in the app.ipynb notebook
• Install missing packages imported in the first cell using magic commands !pip install <package> Note: app.py is the python version of notebook but it's not updated.

• Run the following line to run any component
• python <path to component> --sentence <Input Sentence> --host <corenlp host> --port <corenlp port>
• Eg. python NER.py --sentence "John works for Google." --host "http://localhost" --port 9000