Applications for Predicting Airfare Prices based on Machine Learning Techniques and the Flask Web App

Applications for Predicting Airfare Prices based on Machine Learning Techniques and the Flask Web App

Publication Date : 2024-03-16
Author(s) :

A. Tamizharasi, P. Ezhumalai

           
Article Name :

Applications for Predicting Airfare Prices based on Machine Learning Techniques and the Flask Web App

Abstract :

The flight fare prediction model is meant to receive user inputs from the Flask web application, including departure and arrival cities, trip dates, and airline preferences. The flight price predictor estimates flight fares by comparing the previous prices of flights for the dates and destinations. This paper proposes an applications for predicting airfare prices based on machine learning techniques and the flask web app. This paper uses Machine Learning (ML) techniques, which provide a collection of data accurately and efficiently. Additionally, it is also used to forecast future requirements. The input dataset is initiated by a preprocessing technique, which helps clean the datasets, and the data exploration tries to explore the datasets. ML classification is the next step, which is conducted using the Gradient Boosting Algorithm (GBA). This technique is used in regression and classification tasks. Hyper-parameter tuning in gradient boosting is essential for the overall performance of the machine learning model. The last step in the Flask application is to connect to the database and display the flight price data to users. This paper provides excellent specificity and sensitivity for classifying flight prices. Further provides users with a better understanding of when to purchase tickets and how long they profitable. Overall, the proposed work is developed using the Python Jupyter software. The Specificity Comparison of GBA is 93.5% and Sensitivity Comparison of GBA is 92.5% respectively.

No. of Downloads :

17