🎬 Movie score prediction 🍿

In the world of streaming, services like Netflix rely on data to understand which movies and series users will appreciate. You are a consultant for a start-up that wants to predict a movie’s score based on its metadata.

You have access to a dataset of Netflix movies and series:

• train.csv – movies and series with known scores
• test.csv – new movies and series for which you must predict the score

📊 Dataset

Each row represents a movie title:

• SampleID – unique identifier of the movie
• Title – the name of the movie or series
• Type – content type (SHOW)
• Description – short description of the movie/series
• Year – release year
• Score – only in train.csv, numeric value representing the movie’s score (e.g., critics’ rating)

Your goal is to predict Score for each title in test.csv.

📝 Task (100 points)

Build a machine learning model capable of predicting the numeric value Score for each title in test.csv, using the available columns (Title, Type, Description, Year).

🧮 Evaluation

• The main metric is MAE (Mean Absolute Error):

• MAE ≤ 0.65 → 100 points
• MAE ≥ 2.0 → 0 points
• Intermediate values receive proportional scoring.

📄 Submission File Format

The submission.csv file must contain one row for each title in the test set:

SampleID, Score

where:

• SampleID – the movie identifier from the test set
• Score – the predicted numeric score