You are on page 1of 15

Flight Fare Prediction System

Guided By Assistant Professor : Mr. Sunil Khare

Deepak Jaiswal (1701010051)


Abhishek Kumar (1701010005)
Ayush Raj Singh (1701010043)
Abhishek Kumar Mishra (1701010006)
Index

1. Why we are making this Project


2. What we are doing
3. Algorithm to be used
4. Flowchart
5. Challenges that we have faced
6. What we had done
7. What we need to do in future
Introduction

 As we know that in this era , everyone needs an application by using


which they can get the solution of their problem

 Our Flight Fare Prediction App is a Machine Learning based project


which will predict the Flight Fare depending upon various features
like number of Stoppage , Airlines you want to travel , source ,
Destination ,Departure date , Arrival date , Duration of flight etc
ML model running on

Flow Diagram EC2 instance of AWS

Send Http
Request To our
server by filling
details in app

Return back the predicted


fare of flight
Glance of User Interface
Level-0 DFD
Level -0 DfD

ML model
Use work on Give the
Random
Forest Algo.
Algorithm to be used

 Random Forest Regression :


 Random Forest test the combinations of features also , whereas in Linear
Regression it only work if we manually do , but in Random Forest it do this thing
automatically
 Linear Models have very few parameters , Random Forest a lot more . That means
Random Forest will overfit more easily than a Linear Regression
 Random Forest is a powerful model
Pictorial Representation of Random
forest Regressor
About Random Forest

 A Random Forest is an ensemble technique capable of performing both regression and


classification task.
 By use of multiple decision trees and a technique called Bootstrap and Aggregation,
commonly known as bagging.

 The basic idea behind this is to combine multiple decision trees in determining the final
output rather than relying on individual decision trees.
Software used in this Project

 JUPYTER NOTEBOOK : To perform various Machine Learning techniques


on the Dataset
 PyCharm : used to run server using the Flask Package
 VS Code : used to run the HTML , CSS and JavaScript file for web
deployment
Technology and Tools wise this Project covers :

 Python
 NumPy and Pandas for data cleaning
 Matplotlib for data visualization
 Sklearn for model Building
 Jupyter Notebook , VS code and PyCharm as IDE
 Python flask for Http Server
 HTML , CSS and JavaScript
Diving into the Dataset Problems that we found

 Feature Selection and Feature elimination : which feature should we should


select and which feature to eliminate ?

 How to deal with the String Data type in our Dataset [object data-type] or how
to deal with the categorical features

 The problem of Dummy Variable Trap or we can say that , the problem of
multi-collinearity
Things we have completed

 We are done with our Machine Learning model

 We have created our web user interface

 We have also configured our server


Things need to be Done

 We need to develop our Mobile application [Flutter application] which gonna


run on android as well as IOS

 We need to transfer our Machine Learning model to the cloud or server


running in AWS

You might also like