Spaceship Titanic Prediction Dataset

Table of Contents

• Overview
• Supported Tasks
• Dataset Statistics
• Dataset Structure
• Reference
• License

Overview

The Spaceship Titanic dataset is a synthetic, fictional-disaster dataset of interstellar passenger records. Each row describes a passenger and whether they were transported to another dimension during the anomaly. It is the working data for the Spaceship Titanic Prediction Challenge.

Supported Tasks

Binary classification: predict the Transported label for each passenger.

Dataset Statistics

Property	Value
Total samples	8,693
Training samples	6,085
Test samples	2,608
Features	13
Classes	2 (Transported: True, False)

Dataset Structure

Clone the repository:

git clone [email protected]:AIOZAI/Spaceship_Titanic_Data.git

Or download directly: Spaceship Titanic Prediction Dataset.

Files

File	Rows	Description
train.csv	6,085	Training data with the Transported label
test.csv	2,608	Test data for submission; Transported is withheld

Data Fields

Field	Type	Description
PassengerId	string	Unique passenger ID in the form gggg_pp (gggg = group, pp = number within group).
HomePlanet	string	Planet of residence the passenger departed from.
CryoSleep	boolean	Whether the passenger elected suspended animation for the voyage.
Cabin	string	Cabin assignment as deck/num/side (side P = Port, S = Starboard).
Destination	string	Planet the passenger will debark to.
Age	float	Passenger age.
VIP	boolean	Whether the passenger paid for VIP service.
RoomService	float	Amount billed at the Room Service amenity.
FoodCourt	float	Amount billed at the Food Court amenity.
ShoppingMall	float	Amount billed at the Shopping Mall amenity.
Spa	float	Amount billed at the Spa amenity.
VRDeck	float	Amount billed at the VR Deck amenity.
Name	string	Passenger first and last name.
Transported	boolean	Whether the passenger was transported to another dimension (train.csv only).

Reference

For more information about the dataset, please visit the Kaggle website.

License

Attribution 4.0 International (CC BY 4.0)