Flight Fare Prediction Machine Learning Project with Deployment | Time Series | Project#10

preview_player
Показать описание
🔥 Flight fare prediction is a classical problem of time series forecasting that find trends in past observations to outline the future.

Many popular flight booking websites today, including Google Flights, showcase important insights on: current fair status: high, low or fair; past / upcoming fare trends; and essentially, helps decide the right time to book a flight ticket.

In this project, we are going to build a Python Flight Fare Prediction App, that returns the fare prediction for a given set of travel details, like: departure date, arrival date, departure city, arrival city, stoppages, and the airline carrier.

🔥 Sections:

00:00 Introduction
01:49 Our Plan of Action
05:18 EDA (Feature Engineering)
14:20 Feature Selection
17:41 Model Training
19:30 Predictions on Fresh Data
22:24 Flask Deployment
31:17 Let's talk Machine Learning

🔥 During the course of next ~30mins, we shall discuss:
a. Business use-case for Flight Predictions
c. Feature Engineering on Categorical Variables (using: OneHotEncoding & LabelEncoding)
d. Feature Selection using Sklearn Feature Importance & Variable Inflation Factor (VIF) - for Multicollinearity check
e. Training Fare Prediction - Random Forest Regressor Model
g. Flask Deployment of Project App

🔥 Important Links:

Sentiment Analysis Project (End-to-end) with ML Model Building + Deployment (using Flask):

🔥 Do like, share & subscribe to our channel. Keep in touch:
Рекомендации по теме
Комментарии
Автор

Thanks for your detailed explanations. But it was nice if an airline chosen in departure or source is eliminated from destination drop down list. Of course it is front end work. In case there is a shorter way, pls share it.

shumettefera
Автор

pip install -r .\requirements.txt error

lazeez_corner
Автор

This helped me a lot! Thank you so much

sreevidyabhimagunta
Автор

Hi! Thank you very much for your video. What if we have daily data on multiple flights, airlines, departure dates and routes? Would one hot encoding still be advisable if we have thousands of different airlines and cities?

Thank you very much

andreahenechesierra
Автор

bro i am working same project i want 98.3 accuracy but i used all models but 97 accuracy is coming what will i do to get 98.3 accuracy

NAVEENKODE-ip
Автор

Sir I have a problem in the last step activating the ENV the statement I receive is THE COMMAND NOT FOUND" can you help with that

varshithr
Автор

ModuleNotFoundError : No module named ' flask_cors' please reply

lazeez_corner
Автор

Same Error in all your deployed projects, kindly help

`numpy.distutils` is deprecated since NumPy 1.23.0, as a result
of the deprecation of `distutils` itself. It will be removed for
Python >= 3.12. For older Python versions it will remain present.
It is recommended to use `setuptools < 60.0` for those Python versions.

rohitgupta
Автор

how the prediction is done using features or how much percentage of what feature you are deciding to give a prediction, is there any equation behind can you please tell?

aiswaryalakshmi
Автор

.\ENV\Scripts\activate does not activating can any one help me ??

charan_kamsala
Автор

Please can you characterize Vader as an unsupervised machine learning algorithm. or it is just a dictionary

ahurein
Автор

Brother, is this a Time series dataset ? Can we apply different models like ARIMA, SARIMA, LSTM, MLP. Can we apply these times series models on this dataset of flights ?

faique
Автор

Can I know the over all algorithms used

himadrihere
Автор

sir the problemn is there is no bangalore bcs of one hot encoding in the ui

debojitmandal