Skip to content

Analysis of Gilmore Girls script using Selenium, NetworkX and Steamlit

Notifications You must be signed in to change notification settings

juliakubisa/gilmore-girls-network

Repository files navigation

Gilmore Girls Network

Zrzut ekranu 2024-03-14 o 10 59 19

Data Scraping

First, the list of all characters was acquired from Gilmore Girls fandom wiki using Selenium-Webdriver. Then, also using Seleninum, the scripts from each episodes were scraped from this website. The script dataset contains:

  • character
  • line
  • season

Data cleaning and EDA

Because the data didn't have any missing values, only small spelling mistakes (such as Loreai -> Lorelai) were manually corrected. Then using visualitaion tools (seaborn, matplotlib and NLP libraries (NLTK, re), the simple analysis of the Gilmore Girls script was conducted and placed into a dashboard. Moreover, a wordcloud was created.

Relationships Network

image

The network was created using NetworkX and the more visually pleasing graph was possible with Pyvis Network.

Centrality

There are multiple measures of centrality used in the network anlaysis. Four of them were tested and visually represented in this project:

  • Degree Centrality:
  • Betweenness Centrality:
  • Closeness Centrality:
  • Eigenvector Centrality:

centrality