Fare Intelligence
at your Fingertips

Live demo Objectives

PROBLEM
STATEMENT

OVERVIEW

Predicting railway ticket fares accurately is a complex challenge influenced by factors like travel duration, class preferences, catering services, and distance.

Goals and Objectives

Leverage the IRCTC dataset and Gradient Boosting Regression to develop a precise fare prediction model, providing passengers with transparent and reliable estimates for enhanced travel planning

Responsive

Scalable

Accurate

AboutDataset

The primary source of this dataset is the Indian Railways Catering and Tourism Corporation (IRCTC) website during October-2023. Through web scraping techniques, data was collected, including public train schedules, pricing details, and availability information.

Distance

Chart your fare based on the journey's length.

Duration

Explore how travel time influences ticket pricing

Class

Uncover the impact of seat class on your fare

Catering Service

See how onboard amenities shape ticket costs

METHODOLOGY

Dataset Processing

The dataset is cleaned and preprocessed and the key factors analysing price are taken into account. The dataset is splitted to train and test in 80 : 20 ratio.

Model Training & Exporting

Gradient Boosting Regerssion model is loaded from sklearn library. The model is trained and exported using the pickle module . The process is known as Pickling (serialization)

Evaluation & Metrics

The Accuracy is evaulted for measuring the model's efficeiency. We achived an accuracy of 0.9775 compared to the traditional regression model which had a less score

Integration

The Frontend was built using Bootstrap and handling the requests by Flask framework. The post requests were sent to model by deserializing the earlier saved model. The Output is again sent to frontend

Live Demo
Price Prediction

Experience the excitement of predicting prices in real-time with our live demo. Witness the precision and innovation of our platform at your fingertips.

Catering Service

Dynamic Fare