Gensim Summarization Extension

Python library to extend gensim's summarization library

Features

• Replaced gensim's own regex based sentence splitter with Stanford CoreNLP sentence splitter
• Added functions to get sentence number chosen by the summarizer & replace back sentences using the sentence number
• Added summarizer parameter to give user option to ignore placeholders element inside squarebrackets, i.e. [FORMULA], when calculating sentences scores

Installation

1. Make sure you have Python 3 & setuptools installed
2. Install Stanford CoreNLP
3. Install with pip
   $ pip install gensim_sum_ext
4. To use the library, run Stanford CoreNLP
   $ java -mx4g -cp "*" edu.stanford.nlp.pipeline.StanfordCoreNLPServer -port 9000 -timeout 50000

(Optional) Running Demo Webpage

1. Navigate to example directory
   $ cd example
2. Install pip requirements
   $ pip install -r requirements.txt
3. Run the webserver
   $ python run.py
4. The webpage will be available in 127.0.0.1:5000