GitHub - Roshrini/NLP_question_answering_system_project: NLP Rule based question answering system

1 Branch 0 Tags

Name	Name	Last commit message	Last commit date
Latest commit Roshrini Added files via upload Apr 29, 2016 cd462a9 · Apr 29, 2016 History 2 Commits
developset	developset	Initial commit	Apr 29, 2016
testset1	testset1	Initial commit	Apr 29, 2016
Readme.txt	Readme.txt	Initial commit	Apr 29, 2016
contributors.txt	contributors.txt	Initial commit	Apr 29, 2016
location.txt	location.txt	Initial commit	Apr 29, 2016
location_prepo.txt	location_prepo.txt	Initial commit	Apr 29, 2016
month.txt	month.txt	Initial commit	Apr 29, 2016
names.txt	names.txt	Initial commit	Apr 29, 2016
numberInWords.txt	numberInWords.txt	Initial commit	Apr 29, 2016
occupation.txt	occupation.txt	Initial commit	Apr 29, 2016
preposition.txt	preposition.txt	Initial commit	Apr 29, 2016
qa.py	qa.py	Initial commit	Apr 29, 2016
question-answers.py	question-answers.py	Initial commit	Apr 29, 2016
question_answers1.py	question_answers1.py	Added files via upload	Apr 29, 2016
testset1-inputfile.txt	testset1-inputfile.txt	Initial commit	Apr 29, 2016
time.txt	time.txt	Initial commit	Apr 29, 2016

Repository files navigation

Instructions to run our project on cade:

1) We are using Python3.5 and NLTK package
2) Please run following command and install additional packages.
/usr/local/stow/python/amd64_linux26/python-3.5.0/bin/python3 -m nltk.downloader -d /home/$USERNAME/nltk_data all

Replace $USERNAME with your cade username.

Open our folder. Go to question-answers folder where there are other files.

3) /usr/local/stow/python/amd64_linux26/python-3.5.0/bin/python3 question_answers.py ./developset/input.txt > Response.answer

After running above command, Response.answer file will be generated. This is our output file.

You can calculate accuracy on this file. We got accuracy as 25.89% on developset.