Project - IMDB

The goal in this project is to train a binary classifier to distinguish highly rated movies from low rated movies.

Submissions for this project will be shown on the IMDB Leaderboard.

Training Data

The primary files contain an identifier, as well as title, release, runtime and voting information for several movies. The label column denotes whether a movie was highly rated or not.

train-*.csv

tconst	primaryTitle	originalTitle	startYear	endYear	runtimeMinutes	numVotes	label

In addition, there are two files that contain information about directors and writers of the movies.

directing.json

tconst	director_id

writing.json

tconst	writer_id

Validation & Test Data

We provide validation and test data as input for the submissions. This data has the same format as the training data, but does not contain the corresponding label.

validation_hidden.csv test_hidden.csv

tconst	primaryTitle	originalTitle	startYear	endYear	runtimeMinutes	numVotes