Stanford CoreNLP (http://nlp.stanford.edu/software/corenlp.shtml) HTMLCleaner (http://htmlcleaner.sourceforge.net/) JDOM (http://www.jdom.org/)
- Extract the zip into a Eclipse workspace directory.
- Download the docs directory into the train directory, download at http://cl.ly/2c3q3D3Y0T2K2i0o2B0V
- Download the required libraries from above and add them to the classpath.
Following the model on http://en.wikipedia.org/wiki/Question_answering