https://github.com/ahmedabdalkreem/airline_passenger_satisfaction

In this project, we work in machine learning project that show the satisfacion of customer from the service that have in airline this service make it like this experience or not.
https://github.com/ahmedabdalkreem/airline_passenger_satisfaction

decision-trees gradient-boosting knn logistic-regression machine-learning matplotlib naive-bayes numpy pandas preprocessing python random-forest svm visualization

Last synced: about 21 hours ago
JSON representation

In this project, we work in machine learning project that show the satisfacion of customer from the service that have in airline this service make it like this experience or not.

• Host: GitHub
• URL: https://github.com/ahmedabdalkreem/airline_passenger_satisfaction
• Owner: AhmedAbdAlkreem
• Created: 2023-05-26T00:53:28.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-05-26T23:20:06.000Z (over 1 year ago)
• Last Synced: 2024-01-26T10:40:53.922Z (10 months ago)
• Topics: decision-trees, gradient-boosting, knn, logistic-regression, machine-learning, matplotlib, naive-bayes, numpy, pandas, preprocessing, python, random-forest, svm, visualization
• Language: Jupyter Notebook
• Homepage:
• Size: 1.4 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Airline_Passenger_Satisfaction

In this project, we work in machine learning project that show the satisfacion of customer from the service that have in airline this service make it like this experience or not.

# What the project include?

1) Feature selection and Feature Enginnering.

2) Preprocessing dataset.

3) Train a lot of different model.

4) Change hyperparameter to arrive the best solution.

5) Test the best model and save it to use in real problem.

# Feature selection and Feature Enginnering:

In this step we try to know which the column is important in the data to deal with it and the coulumn that not import to drop it from dataset to avoid 

overfitting in dataset when train data and splliting dataset to input and output.

# Read and preprocessing dataset:

In this step read dataset and know what a problem in dataset to try to solve it such as delete column that have missing values in dataset then encoding

binary and ordinary data and one_hot_encoding in nominal data before data be numbers can be ready to be input of model but before this step we make 

normalization to dataset to be in the same range then finished this step.

# Train a lot of different model:

In this step we try to use different model and train and validation all the model to arrive the best model solve the problem and give me the best 

accuracy without overfitting in dataset and this is the goal that try to arrive to make it.

# Change hyperparameter to arrive the best solution:

In this step we try to change the hyperparameter of all model to arrive the best parameter that make model arrive the best accuracy the best hyperparameters 

called optimal solution and this is the goal in this step to try to improve all model to arrive the best hyperparameters in model and this inhance 

accuracy of model and make best accuracy.

# Test the best model and sace it to use in real problem:

This is the last step of this project take the best model that make best accuracy in train and validation and test it then save the model to use it

in any problem of real life.

# What will learn before this project?

1) Read dataset and know problem in dataset.

2) Deal with the missing values in dataset.

3) Encoding dataset with manual incoding or one_hot_incoding.

4) Make normaliztion to make data in the same scale.

5) Train, validation and test Model.

6) Learn how to arrive the best hyperparameter.

7) Make visualiztion to dataset and know the important columns.

8) Save the model to use it later.