IMDB

Stats on movie roles by age and gender

• In the "imdb 0 donnees.ipynb" notebook we select data from imdb database to build a database of roles in movies ("principals" in ibdb parlance) which is written in a imdb_dbr.csv file please note that for the notebbok to be run successfully one need to first download the imdb data (https://www.imdb.com/interfaces/)
• In "imdb 1 roles.ipynb" we use the data from compute stats on the imdb_dbr.csv files to compute various stats on roles the pdf files presents (in french) the results obtained
• In "imdb 2 carrieres.ipynb" we prepare data for use in a model. To this end we form a cohort of actress and actor active between 2008 ans 2012
• In "imdb 3 modele carrieres-USA.ipynb" we (not quite successfully) try to build a model to predict wether a given actor/actress will got new role in the following years. Given previous results, the non-significance of gender is upsetting. There must be a mistake.