papawcoin / airline-price-prediction

Airfare Prediction using XGBoost and Neural Network (Keras with a Tensorflow backend)

Airline-Price-Prediction

The advantages of travelling by airplanes have increased overtime. The industry tries to make the ticket fare reasonable as well as to make profit out of it. Airline industry has lots of dynamic factors affecting them in a day today operation. It’s one of the highly sophisticated industry which aims at making revenue. The purpose of this study is to better analyze the features that affect airfare and develop and tune models to predict the airfare well in advance. We used XGBoost and Keras with a Tensorflow backend Neural Network, two state of the art prediction models for our study and used various machine learning tasks to achieve the best performance for our task.