Classification Airline Passenger Recommendations

Data includes airline reviews from 2006 to 2019 for popular airlines around the world with multiple choice and free text questions. Data is scraped in Spring 2019. The main objective is to predict whether passengers will refer the airline to their friends.

Project Summary

The airline industry plays a crucial role in modern transportation, with numerous airlines serving various routes worldwide. To make informed decisions in this highly competitive industry, airlines and stakeholders rely on data-driven insights. Machine learning models are indispensable tools in this regard, allowing for the classification of airlines based on different criteria. This document outlines the development and implementation of an airline classification machine learning model.

This is where machine learning can play a vital role. By using historical customer data, a machine learning model can identify patterns and correlations that indicate a high likelihood of referral. This information can then be used by airlines to target specific customers with personalized marketing campaigns or incentives, increasing the chances of referral and promoting growth.

In conclusion, a machine learning model that predicts the likelihood of referral can provide valuable insights for airlines looking to enhance their customer satisfaction and drive growth through word-of-mouth referrals.

Problem statement

The goal of this machine learning project is to classify airlines into categories based on certain features or attributes. Classification can serve multiple purposes, such as identifying potential partners for codeshare agreements, assisting in pricing strategies, or aiding in market analysis. In this project, We will be exploring if flyers would recommend the airline to their friends and families, based on their travel experience,reviews and ratings.

There are few problems that we are looking in this project:

• Develop a classification model to categorize airlines based on the likelihood of customers recommending them to friends and family.

• Recognize the pivotal role of customer satisfaction and referrals in the growth and success of airlines.

• Enable airlines to strategically utilize customer referral information for codeshare agreements, pricing strategies, and market analysis.

• Identify customers likely to refer the airline, a task complicated by the diverse factors influencing satisfaction and referrals.

• Assess the model's capability to provide actionable insights for airlines to tailor services, improve customer satisfaction, and enhance brand reputation.

This problem statement outlines the key objectives, challenges, and considerations for developing a classification model to predict customer referrals in the airline industry.

Objective

The primary objective is to deploy a robust classifier model that accurately predicts user recommendations, empowering airlines to strategize effectively, understand user behavior, optimize services, and align business strategies with financial objectives.

Steps involved are:-

1. Know Your Data

2. Understanding your Data

3. Data Cleaning

4. Data Manipulation

5. Data Visualization

6. Hypothesis Testing

7. Feature Engineering & Data Pre-processing

8. Data Splitting and ML Model Implementation:-

        a.Decision Tress
      
        b.K-Nearest Neighbour
      
        c.Support Vector Machine
      
        d.Random Forest Classifier
   

9. Feature Importance

10. Deployment File

          a.Pickle
        
          b.Joblib