Forecasting Taxi Cancellation of Online Booking App

Abstract

In this project we are going to attempt to predict doable cancellations of cab booking by the client
exploitation knowledge obtained from the corporate. Our goal is to cut back the price incurred by the corporate as a
results of cab cancellations created by the client. By predicting doable cancellations associate hour before the pickup
time, company are going to be higher able to manage its vendors and drivers by providing them with up-to-date info
regarding client cancellations and cut back the price incurred from causing a cab to a booking location that has been off
by the client. Accurate prediction of client cancellations can cause a discount in company prices. Our knowledge
analysis model used effective strategies to investigate the information like Naïve Bayes. The accuracy of the model as
well as the ultimate business goal of reducing price for the corporate was wont to settle the model for the prediction. The
model that we have a tendency to chosen within the finish was Naïve Bayes. Not solely will the model have associate
overall low error rate, however additionally the value incurred by the corporate exploitation this model is that the
lowest. Our recommendation includes running the model in real time on associate hourly basis for all pickup times, that
square measure among associate hour’s time. The model can flag all possible booking cancellations on net application.
By exploitation the model for predicting doable client cancellations, the corporate can with success cut back the price
incurred from causing a cab to a pickup location wherever the client isn't gif