readme.txt

qa.txt is our original dataset.
generateCorpus.py will generate corpus accordingly

match_evaluation.py is a customized module in the following code

For task 2, run:
python3 Task2.py 50corpus_copy.txt
Then read the screen output, processing has been divided and noted. Matched results can be found under:
"*********** start matching results in corpus *************
"the best 10 matching results"

For task 3, run:
python2.7 Task3.py 50corpus_copy.txt
Then read the screen output as formated the same in task 2

For stanford parser, please put .jar files under the folder called "StanfordParser". Due to the size limit, we don't have the files here and you'll need to download them.They are:
stanford-parser-3.9.1-models.jar
stanford-parser.jar

All the rest details please refer to our report. Thank you!