In this project, information retrievel and text mining is used to extract information from the Cornelia Funke books Inkheart. In this project an attempt is made to extract information about the characters in the books and about the (geographical) events throughout the books.
-
Who are the characters in the books, can we use named entity recognition to find the different characters, and is it possible to extract certain attributes about the characters. Attributes such as whether they are from inside of a book or from the ‘real’ world.
-
Who do the characters interact with the most during the three books? Is is possible to create a relation mapping between the different characters to see who are the main protagonists, and who can be seen as ‘side-kicks’
-
Where do the characters go throughout the story. As explained above, the topology2 of the characters changes quite drastically as the characters move in- and outside of fictional worlds. Is the information retrieval system able to differentiate these universe changes and model the movement of the characters throughout the three books?
In order to be able to run all the used modules in this project, run:
pip install requirements.txt
This is a project for the IRTM Course taught at Maastricht University's Artificial Intelligence.