MLP Project

May 2023

Problem statement

In this dataset each record represents a movie-review pair with movie title, description, genres, duration, director, actors, users' ratings, review text, reviewer name, etc. Your task is to build an ML model to predict sentiment of the review text.

Dataset

Files:

• train.csv - the training set containing the review sentiment along with other features.
• test.csv - the test set, has review features, but no sentiment column, since it is the target.
• movies.csv - the file with metadata on movies.

Dataset

Columns in train.csv:

• movieid - named id of the movie
• sentiment - indicating "POSITIVE" or "NEGATIVE", which is the target.
• Other columns are self explanatory.

Columns in movies.csv:

movieid, title, audienceScore, rating, ratingContents, releaseDateTheaters, releaseDateStreaming, runtimeMinutes, genre, originalLanguage, director, boxOffice, distributor, soundType

The column names are self explanatory.