Oscar nominations and winners are usual the resultant of nominations and winners of less prestigious awards (e.g. Golden Globes, BAFTA, Critics Choice, industry awards). Supervised machine learning models are used to predict 2021 Academy Awards nominations and winners based on 5 previous years' nominations.
Nominations - summary.pdf contains a summarized report about predictions of nominations.
Winners - summary.pdf contains a summarized report about predictions of winners.
Oscars-TrainingData.csv contains all data about 5 last years, including nominations for Oscars, Golden Globes and so on (234 movies x 100 categories). For every movie there is a dedicated number (mostly 0 or 1, sometimes 2) that indicates how many nominations did the film achieve in a given category. This is a training data for the model.
Oscars-TestingData.csv contains the same data about 2021 movies except for Oscars nominations (61 movies x 84 categories).
OscarsWinners-TrainingData.csv contains all data from Oscars-TrainingData and additionaly info about winners of particular Oscar categories.
OscarsWinners-TrainingData.csv contains all data from Oscars-TestingData and additionaly info about nominations for particular Oscar categories.
All the collected data is based on filmweb.pl.
OscarsNominations.py is an executable file which enables a user to pick a category and see a predicted nominees.
OscarsWinners.py is an executable file which enables a user to pick a category and see a predicted winner.
preview.png and preview2.png are views from Python Console while executing OscarsNominations.py.