Chamber of Relations: A Harry Potter Network (Work in Progress...)
Unveiling the Wizarding Web: Creating a Network of Harry Potter Characters
Dive into the magical realm of J.K. Rowling's Harry Potter series, where spells are cast, friendships are forged, and the intricacies of character relationships weave a tapestry of captivating narratives. In this series, we embark on a journey to unravel the secrets of the wizarding world by creating a network of interactions between its beloved characters.
The Magical Blueprint
Our journey begins with a meticulous exploration of the characters in the Harry Potter series and the chapters in which they first make their enchanting appearances. The foundation of our character interaction network lies in the data we've compiled, and you can find the magical ingredients on our GitHub repository at nikhil-ravi/Harry-Potter-Network.
All of the network data is available on our GitHub repository at nikhil-ravi/harry-potter-interactions.
1. Web Scraping: Unveiling the Characters
The first enchanting spell we cast involves scraping the pages of the Harry Potter books to gather information on characters and the chapters they grace with their presence. Our Web Scraping page takes you through the intricate dance of data retrieval, bringing forth the names and initial appearances of characters.
2. Resolving the Mysteries of Coreferences
The magical prose of J.K. Rowling often weaves pronouns, nicknames, and aliases into the narrative, creating an intricate tapestry of coreferences. To unravel this linguistic magic, we turn to FastCoref and SpaCy, our trusty companions in resolving these mysteries. The Entity Resolution page unfolds the process, ensuring that each character is distinctly identified in the magical text.
3. Summoning Entities with SpaCy's Matcher
With coreferences resolved, our next step involves summoning entities from the text using SpaCy's rule-based entity matcher. This ensures that each character is granted a distinct identity, laying the groundwork for meaningful interactions.
4. Extracting Interactions: A Magical Connection
The heart of our wizarding network lies in the interactions between characters. We employ a spellbinding approach, extracting interactions whenever two characters are mentioned within a certain number of words of each other. This process, detailed on subsequent pages, brings to light the dynamic relationships within the wizarding world.
Join us on this magical journey as we uncover the connections, friendships, and rivalries that define the Harry Potter series. Each page in this section delves deeper into the intricacies of our methodology, allowing you to witness the creation of a network that mirrors the enchanting complexity of J.K. Rowling's masterpiece.