Titanic Survival Prediction using TensorFlow

Overview:

• This project aims to predict the likelihood of a passenger surviving the Titanic shipwreck
• Preprocessing techniques to deal with both categorical and numerical data types
• Model is trained on features such as age, sex, class, etc using a Linear Classifier provided by TensorFlow
• Performance is evaluated using the testing data, and the accuracy metric is printed as an outcome

Requirements:

• Python 3.x
• TensorFlow
• Pandas
• Numpy
• Matplotlib

Dataset:

• Age
• Sex
• Number of siblings and spouses aboard
• Number of parents and children aboard
• Passenger class
• Deck
• Embarkation town
• Alone status
• The label for prediction is whether the passenger survived or not
• The dataset is split into training (train.csv) and testing (eval.csv) data

Installation:

1. Clone the repo:

https://github.com/williamlin6803/Survival.git

2. Run pip install -r requirements.txt to install dependencies
3. Execute the Python script: python titanic_survival_prediction.py

Results:

• The model achieves an accuracy of 78% on the test data

Future Plans:

• Implement more sophisticated models that uses other Machine Learning Techniques such as Random Forests or Neural Networks for better accuracy