scisumm

Single Document Summarization of Scientific Documents

Instructions:

Run the requirements.txt through pip pip install -r requirements.txt
Additionally, nltk data needs to be downloaded without which the scripts wont run succesfully. run nltk.download() inside a python shell currently, the stopwords from the corpus are required.
Download ROUGE and put the folder RELEASE inside scisumm/lib/ROUGE
Change the address of the base directory in scr/Config.py

Name	Name	Last commit message	Last commit date
Latest commit junkmechanic Work towards including n-grams Apr 14, 2014 ee556d3 · Apr 14, 2014 History 74 Commits
data	data	1st Iteration Analysis	Mar 10, 2014
demo	demo	Added docs. summarize by section.	Mar 14, 2014
docs	docs	Changes before submission	Apr 3, 2014
lib/Stanford-Parser	lib/Stanford-Parser	After testing with different levels of weight	Feb 23, 2014
src	src	Work towards including n-grams	Apr 14, 2014
.gitignore	.gitignore	After classification using tfisf and counts	Jan 28, 2014
README.md	README.md	sumseciton2 for TextRank on individual sections	Dec 9, 2013
requirements.txt	requirements.txt	Implemented TextRank. Improvements to Document and Sentence classes	Nov 4, 2013

Provide feedback