Titanic - Survival Prediction

Project Overview

This is the Titanic ML competition from Kaggle
According to Kaggle, ..."it's the best, first challenge for you to dive into ML competitions and familiarize yourself with how the Kaggle platform works."

The competition is simple: use machine learning to create a model that predicts which passengers survived the Titanic shipwreck.

Installation and Setup

Codes and Resources Used

• Editor Used: Jupyter Lab
• Python Version: 3.11.7

Python Packages Used

• General Purpose: warnings
• Data Manipulation: pandas, numpy
• Data Visualization: seaborn, matplotlib
• Machine Learning: scikit, sklearn

Data

Source Data

• train.csv
  • from Kaggle's Titanic web page
• test.csv
  • from Kaggle's Titanic web page

Data Acquisition

Simply navigate to the datasets using either link above and download the files.

Data Preprocessing

• titanic_combined_dataset_cleaned_processed.csv
  • I combined both train and test to clean and perform EDA
• train.cleaned.csv
  • modified train.csv file ready for modeling
• test.cleaned.csv
  • modified test.csv file ready for modeling

Code structure

├── data
│   ├── train.csv
│   ├── test.csv
│   ├── titanic_cleaned_processed.csv (file after I prepared data for modeling)
├── README.md
├── titanic_exploration.ipynb
└── .gitignore 

Results and evaluation

I was 12,910 with a 0.76076 Score
This was my first ever solo exploration, I supposed it could have been worse :)

Future work

I may revisit this in the future and try again after I have more experience with competitions.

Acknowledgments/References

Hackers Realm YouTube channel did a video on this competition and I cross-referenced/checked my work against this video.

License

Subject to Competition Rules