eshannaik / Titanic

Predicting if a person will survive in the Titanic incident or not using different models

Titanic

This project was done on Google Colab.

The dataset for the titanic project can be got from https://www.kaggle.com/c/titanic/data.

The dataset for the old titanic.py is https://www.kaggle.com/heptapod/titanic

---

The main objective of this project is to predict if a person will survive during the titanic incident.

The different models that have been used to try to predict this are -

ANN
SVM
Logistic Regressing using GridSearchCV
Random Forest

The different libraries that have been used are :

Pandas
Numpy
Matplotlib
Seaborn
Sklearn
Keras

The accuracies achieved for the different models are :

ANN : 74.55 %
SVM : 85.24 %
Logistic Regression : 84.08 %
Random Forest : 85.24%

As we can see Light GBM has the highest accuracy which is 97.937%