Augmented Novels

Read novels without effort!

Project Structure

data/: contains books and datasets legacy/: contains legacy code for named entity and character description www/: folder to contain the Augmented Novel

Dependencies

Python 2.7
NLTK
numpy and scipy
scikit-learn (only for weight.py)

Generate the Augmented Novel and the interactions graph

Open main.py and change variable txt_file to open the desired book.
Adjust max_chars to control the number of characters in the generated graph
Adjust parameter smallest of draw_graphviz to scale the generated graph
Run main.py

It will generate www/peopleA.html the Augmented Novel with dialog segmentation and character detection. It will also generate www/graph.gv which you can open with GraphViz Neato to generate the interactions graph.

Generate Character profiles

Open legacy/main.py and change variable filetoread to open the desired book.
Run legacy/main.py

It will print word associations with each detected character.

Highlight Important Sentences

Open weight.py and change the book file
Adjust num_segments and constants as desired
Run weight.py