QuickReaderV4

A text summarisation algorithm and interface

Usage

To run the code, python main.py. When you find an article or document to summarize, copy the text, paste it into the white box at the bottom of the interface, and press enter.

Inside TextBox.py, you can set values associated with the colour scheme and font.

Inside Summarizer.py are the algorithms for the text summarization.

When using the program, the following operations are available:

• update (button): after changing the coherence or independence params (but not length), press update to recalculate the summary
• less/more (button): use this to toggle between showing only the summary or showing the entire document with the summary highlighted
• length (slider): this changes how many sentences are extracted for the summary. When set to 0, the entire document is shown.
• coherence (slider): this changes the weight of coherence (i.e. how important it is for consecutive sentences in the summary to be related)
• independence (slider): this changes the weight of independence (i.e. how important it is to include sentences in the summary that introduce new content/do not depend on previous sentences for comprehension)

Dependencies

This project is written in Python2.7 with Tkinter for the GUI. It requires the following non-default libraries:

• nltk
• numpy, scipy
• ScrolledText