Use of Data Science to Predict Survival of Titanic Passengerss


  • Leonard Garrison
  • Amanda Carpenter
  • Jon McClung
  • Jeremy Chosie


The fields of data science and machine learning have experienced unprecedented growth over the previous decades. The demand for accurate and practical techniques to analyze and learn from large datasets will likely continue to grow in the future. We discuss some of the opportunities and challenges provided by common machine-learning techniques. In particular, we study the case of the famous Titanic disaster, which resulted in the loss of one of the finest luxury cruise ships of the era and the lives of 1,517 passengers and crew.

Using standard machine-learning techniques, we attempt to predict the survival of its victims based on data including age, gender, cabin number, and so on, and analyze the effectiveness and practicality of these techniques. The exercise demonstrates the merit of an exploratory approach for study. We discuss the scalability and generality of our approach. Our results provide a solid foundation for understanding the utility of data science as a means to predict future outcomes.





Computer Science