Skip to content
Scraped data off Wikipedia, Boxofficemojo, RT to create a "cult index". Predict a movie's "cult index" score by building a model on pre-release information.
OpenEdge ABL Jupyter Notebook Python
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
.ipynb_checkpoints
data
figures
misc
pickles
src
BOMOJOmovies.py
EDA.py
README.md
Untitled.ipynb
predict cult movie status.key
predict_cult_movie_status.pdf
scrapingTestRun.ipynb
wikiMovies.py
workspace cult movies GENERATE TRAINING TEST.ipynb
workspace cult movies categoricals cleanup.ipynb
workspace cult movies numerical predictors eda.ipynb
workspace cult movies.ipynb

README.md

Can Pre-Release Information Predict Movie Cult Status

Started out as a classification problem, but want to treat it as a Linear Regression problem.

  • Step 1: Create a "Cult Index" that's calculated from post-release information,
    ++ Movie's Cult score is rewarded for high lifetime gross revenue, and penalized for high opening weekend and for high number of theaters released in. (still figuring out an ideal index).
    ++ To-do! Remove outliers
    ++ To-do! Adjust lifetime revenue for inflation!! (haven't done this yet)
    ++ To-do! Id more cult movies! Online lists are very subjective and insufficient.

A Note on the Subjective Nature of Cult Status

As you may guess, there is a lot of debate around what passes for a cult movie. Per the literature, there are two broad categories that these can be split in:

Inclusive definitions allow for major studio productions, especially box office bombs, while exclusive definitions focus more on obscure, transgressive films shunned by the mainstream. The difficulty in defining the term and subjectivity of what qualifies as a cult film mirrorclassificatory disputes about art.

Preliminary Cult Index Based Top Cult Movies

Preliminary top cult movies shown by index

You can’t perform that action at this time.