A set of machine learning models developed for the titanic survival dataset
This repo is my introductory project to machine learning. It consists of various machine learning models I developed using scikit-learn to predict which passengers survived the tragedy of the sinking of the Titanic. The data was provided by Kaggle, a data science company that has established an online community of data scientists and machine learners. Using their data set, I used some common classification-based algorithms to produce and fine tune my models
These models are based on simpler machine learning algorithms, including:
- K-Nearest Neighbours
- Logistic Regression
- Support Vector Machine
- Decision Tree
Most of the pre-processing and data analysis has been done using Numpy and Pandas to make logical assessments on what to do with missing data and parameter tuning
The models have been trained and tested models using Kaggle's provided data set, measuring accuracy and performance before submitting to test against the competition's undisclosed test set
To learn more about the competition and download the dataset for yourself, visit the website here