CAPSTONE PROJECT

REVIEW 1 REPORT

AIRPRED-An airline ticket price prediction web app using Machine

Learning

DONE BY: Satkar Acharya 19BCE2521

Under The Guidance Of : Dr.Narayanan Prasanth

INTRODUCTION:
The tourism industry is constantly evolving, resulting in an increase in
the number of travelers each year. One of the most sophisticated
industries within tourism is the airline industry, known for its complex
pricing strategies. Flight prices have become increasingly unpredictable,
with ticket prices fluctuating frequently. Customers are looking for the
best deals on their tickets, while airline companies aim to maintain high
revenue. However, with the use of technology, it is possible to reduce
the uncertainty of flight prices. By utilizing machine learning techniques,
we can predict flight prices more efficiently.

PROBLEM STATEMENT:
Currently, the flight ticket buying system is to purchase a ticket many
days prior to flight takeoff so as to stay away from the effect of high
charges.
The airlines use complex policies of Revenue Management for execution
of distinctive evaluating systems. Due to this, system modifies the
charge according to the time of day, the season, and special occasions
on subsequent pages.
Customers look to buy cheaper tickets way ahead of departure but end
up paying more.

OBJECTIVES:
Analyze different machine learning algorithms to predict the best
accuracy and use it to predict the price.
Classifying and preforming analysis on the dataset. Data cleaning and
encoding method will be used to classify.
Build the web app to provide user interface to choose flight fare on
basis of cheap tickets.
Development to integrate the components of front end to machine
learning models using API calls.
 Build an efficient robust website integrating the components of
machine learning as well as web application.

LITERATURE REVIEW:
Authors Title Source summary reference

K. Tziridis,Th. Airfare 2017 25th this research [1]

Kalampokas prices European Signal paper details a
,G. A. prediction Processing set of features
Papakostas,K. using Conference according to
I. machine which a typical
Diamantaras learning flight ticket
techniques may be
affected.these
features are
applied to eight
ML models and
the accuracy of
each was
taken.Bagging
Regression
Tree Model
showed the
best accuracy.

Supriya Predicting INTERNATIONAL in this study, [2]

Rajankar, The Price JOURNAL OF data for a
Neha Of A Flight SCIENTIFIC & certain air
Sakharkar, Ticket With TECHNOLOGY route,
Omprakash The Use Of RESEARCH comprising
Rajankar Machine VOLUME 8, ISSUE features like
Learning 12, DECEMBER departure time,
Algorithms 2019 arrival time,
and airways for
a specific
period,have
been collected
to determine
the minimum
 airfare.machine
learning models
have been
applied.the
paper
concludes by
saying that the
results would
be more
accurate if the
availability of
seats was also
there in the
data.

J. Santos Optimal Journal of Air In this study, [3]

Domínguez purchase Transport the authors
Menchero, timing in Management examined
Javier Rivera the airline ,August 2014 airfares for
,Emilio Torres market flights from
Madrid to
London,
Frankfurt,New
York, and Paris
during a two-
month period,
accounting for
up to 30-day in
advance ticket
purchases.
They
discovered that
the consumer
had 18 days
before
departure to
buy a ticket
without
suffering a
major financial
loss.
Neel Bhosale, Flight Fare International summary:the [4]
Pranav Gole, Prediction journal for researchers
Hrutuja System research ,may found that
Handore, Priti Using 2022 prices of flights
Lakde, Machine are highly
Gajanan Learning sensitive to
Arsalwad date and time
of departure
and if the
departure date
is a holiday.

Viet Hoang An airfare 2018 The authors of [5]

Vu, Quang prediction International this paper have
Tran Minh, model for Conference on put forward a
Phu H. Phung developing Information new method to
markets Networking assist buyers in
forecasting
airfare prices
without relying
on information
from airlines.
The research
shows that this
proposed
model can
predict the
changes in
airfare prices,
even up to the
day of
departure, by
using publicly
available
airfare data
found online.
However, it
should be noted
that this model
may not be able
 to take into
account certain
important
factors, such as
the number of
unsold seats on
flights.

SYSTEM ARCHITECTURE:

Flask Framework will be used to build the application, will enable us to

create API endpoints like GET and POST to manage tasks related to loading
and displaying data on the front-end.
The frontend part of the web application will be built using HTML and CSS
using the bootstrap framework as well.
The backend service will receive the input data and use a model to predict
the outcome. The user will then see the expected value shown on the front
end.

REFERENCES:
[1]https://www.eurasip.org/Proceedings/Eusipco/Eusipco2017/papers/157
0348051.pdf
[2] http://www.ijstr.org/final-print/dec2019/Predicting-The-Price-Of-A-
Flight-Ticket-With-The-Use-Of-Machine-Learning-Algorithms.pdf
[3]https://www.sciencedirect.com/science/article/abs/pii/S0969699714000
842
[4] https://www.ijraset.com/research-paper/flight-fare-prediction-system-
using-ml
[5]https://www.researchgate.net/publication/324727116_An_airfare_predi
ction_model_for_developing_markets