Skip to content

ML used to solve spaceship passenger disappearance due to spacetime anomaly. Filled missing data, created new features, encoded categorical columns, and trained 4 classifiers (KNN, Decision Tree, Gradient Boosting, and Random Forest). Gradient Boosting had highest accuracy (.8076). GitHub project showcasing ML in problem-solving.

Notifications You must be signed in to change notification settings

jett512/Classification-Models-for-Kaggle-Competition

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Classification-Models-for-Kaggle-Competition

To set up the notebook, make sure that the data folder in included in the directory. Each cell can be ran separately, or ran all together.

In this problem, we are trying to figure out the cause behind the disappearance of passengers on a spaceship through a spacetime anomaly. To do this, we employed various machine learning techniques. The first step was to fill missing values in the dataset and then create new features while dropping some unnecessary ones. Categorical columns were encoded by creating dummy variables for each unique category and normalizing the dataset using the Normalizer() function. Finally, we trained four different machine learning classifiers: KNN, Decision Tree, Gradient Boosting, and Random Forest. The results showed that the Gradient Boosting classifier had the highest classification accuracy of .8076, indicating that it was the best model for this problem. This project demonstrated the effectiveness of using machine learning in solving complex problems by applying data preprocessing techniques and trying different models to identify the most suitable one.

About

ML used to solve spaceship passenger disappearance due to spacetime anomaly. Filled missing data, created new features, encoded categorical columns, and trained 4 classifiers (KNN, Decision Tree, Gradient Boosting, and Random Forest). Gradient Boosting had highest accuracy (.8076). GitHub project showcasing ML in problem-solving.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published