Scribd
Upload
76 views12 pages

Flight Price Prediction Document

This document outlines a project to predict flight prices using machine learning models. The goal is to build a model that can predict flight prices based on factors like booking date, flight duration, departure and arrival times. The project involves collecting flight fare data from websites, cleaning the data, exploring relationships between variables, and using algorithms like linear regression and random forest regression to predict prices. Key metrics like RMSE and R2 score will be used to evaluate model performance. Visualizations will provide insights into price distributions and the impact of different features. The results will help understand which factors most influence prices and how well machine learning can predict them.

Uploaded by

Kushal Tirukovela
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
76 views12 pages

Flight Price Prediction Document

This document outlines a project to predict flight prices using machine learning models. The goal is to build a model that can predict flight prices based on factors like booking date, flight duration, departure and arrival times. The project involves collecting flight fare data from websites, cleaning the data, exploring relationships between variables, and using algorithms like linear regression and random forest regression to predict prices. Key metrics like RMSE and R2 score will be used to evaluate model performance. Visualizations will provide insights into price distributions and the impact of different features. The results will help understand which factors most influence prices and how well machine learning can predict them.

Uploaded by

Kushal Tirukovela
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 12

FLIGHT PRICE PREDICTION

Submitted by:
ADARSHKUMAR
ACKNOWLEDGMENT
This includes mentioning of all the references, research papers, data
sources, professionals and other resources that helped you and guided
you in completion of the project.

I have referred most of the things data trained class notes and some
Machine learning articles from Towards Data science
Selenium issues & data cleaning -- Stack overflow
INTRODUCTION

 Business Problem Framing


Describe the business problem and how this problem can be related
to the real world.
Anyone who has booked a flight ticket knows how unexpectedly the
prices vary. The cheapest available ticket on a given flight gets more
and less expensive over time. This usually happens as an attempt to
maximize revenue based on - 1. Time of purchase patterns (making
sure last-minute purchases are expensive) 2. Keeping the flight as
full as they want it (raising prices on a flight which is filling up in
order to reduce sales and hold back inventory for those expensive
last-minute expensive purchases) So, you have to work on a project
where you collect data of flight fares with other features and work
to make a model to predict fares of flights.

 Conceptual Background of the Domain Problem


Describe the domain related concepts that you think will be useful
for better understanding of the project.
Airline and transportation

 Review of Literature
This is a comprehensive summary of the research done on the
topic. The review should enumerate, describe, summarize, evaluate
and clarify the research done.

 Motivation for the Problem Undertaken


Describe your objective behind to make this project, this domain
and what is the motivation behind.
Analytical Problem Framing
 Mathematical/ Analytical Modeling of the Problem
Describe the mathematical, statistical and analytics modelling done
during this project along with the proper justification.

 Data Sources and their formats


What are the data sources, their origins, their formats and other
details that you find necessary? They can be described here.
Provide a proper data description. You can also add a snapshot of
the data.
Data collection done through scraping different airline websites,
like Yatra.com, EsayMyTrip.com
Data stored in .CSV file format

 Data Preprocessing Done


What were the steps followed for the cleaning of the data? What
were the assumptions done and what were the next actions steps
over that?
We have handled null and missing values , corrected the data types
and format of the independent variables in data cleaning steps
In Data Pre-processing we have checked heat map of correlation
matrix to identify the relation between independent variable and
target variable
Handled the categorical variables using dummy variables
We used feature selection identifying the correct independent
variable to build model
 Data Inputs- Logic- Output Relationships
Describe the relationship behind the data input, its format, the logic
in between and the output. Describe how the input affects the
output.
The problem belongs to the Airline domain so booking ticket date of
the flight and date of journey is more important, also flight duration
is more impact on the out put
Any change in the flight duration and ticket booking and date of
journey change in output (flight ticket price)

 State the set of assumptions (if any) related to the


problem under consideration
Here, you can describe any presumptions taken by you.

 Hardware and Software Requirements and Tools Used


Listing down the hardware and software requirements along with
the tools, libraries and packages used. Describe all the software
tools used along with a detailed description of tasks done with
those tools.
Hardware: Sony vaio laptop ,i5 processor , 4GB Ram
Software : window 10 , Jupyter notebook
Model/s Development and Evaluation

 Identification of possible problem-solving approaches


(methods)
Describe the approaches you followed, both statistical and
analytical, for solving of this problem.

 Testing of Identified Approaches (Algorithms)


Listing down all the algorithms used for the training and testing.

 Run and Evaluate selected models


Describe all the algorithms used along with the snapshot of their
code and what were the results observed over different evaluation
metrics.
 Key Metrics for success in solving problem under
consideration
What were the key metrics used along with justification for using it?
You may also include statistical metrics used if any.
 Visualizations
Mention all the plots made along with their pictures and what were
the inferences and observations obtained from those. Describe
them in detail.
Observation: Airline vistara have highest number of flights

Observation : Data set contains the highest number of flight data


from Bangalore
Catplot:

Observation: Air India have highest Air fare compare to other


airlines and cheapest airline Air aisa
Observation : Correlation of the heat map , Arrival hour and
dur_hours have positive relation between Feature variable and
target variables

Observation: Highly Price distribution happened between 5000 to


10000 Rupees

Fig: Price and Hour scatter plot


Fig: Price and Day scatter plot

Observation: Bar plots show the clearly feature variables Day and
stop ,Dep_hour, dur_hour, Arr_hour have high correlation
between the target variables
Fig: Test score of the y_test, and y_pred

Fig: Scatter plot of y_test and y_pred

If different platforms were used, mention that as well.

 Interpretation of the Results


Give a summary of what results were interpreted from the
visualizations, preprocessing and modelling.

CONCLUSION
 Key Findings and Conclusions of the Study
Describe the key findings, inferences, observations from the whole
problem.

 Learning Outcomes of the Study in respect of Data


Science
List down your learnings obtained about the power of visualization,
data cleaning and various algorithms used. You can describe which
algorithm works best in which situation and what challenges you
faced while working on this project and how did you overcome
that.

 Limitations of this work and Scope for Future Work


What are the limitations of this solution provided, the future scope?
What all steps/techniques can be followed to further extend this
study and improve the results.

You might also like