This is a classification supervised machine learning project completed as part of Project 3 of the Metis Data Science Bootcamp (Singapore). In about 3 weeks, our instructor took us through a whirlwind tour of SQL, AWS, various classification techniques and how to deploy our machine learning model on a Flask app. In this blog, let me take you through the model that I have built based on a flight survey data and demonstrate the business value that it can potentially create.

1. Background

Airline businesses around the world are decimated by Covid-19 as most international air travel has been grounded. Among the hardest hit might be Singapore Airlines, which operates zero domestic flight in its island home nation. In fact, some airlines such as Thai Airways have already filed for bankruptcy. Nonetheless, once the storm is over, demand for air travel is expected to surge as people rush back for overseas holidays. What can airlines prepare to give themselves a competitive edge when the crowd finally arrives? To answer this business problem, a classification model is created from the flight satisfaction survey data from Kaggle to identify the critical factors that lead to customer satisfaction.

#machine-learning #random-forest #data-science #airlines #classification