Predicting airline customer satisfaction: Exploring key factors influencing customer experience and using the learnings to predict the customer satisfactory response.
Customer satisfaction is a cornerstone of success in the airline industry. This project aims to explore and analyze the factors that influence airline customer satisfaction. By identifying key variables, we seek to offer actionable insights to enhance customer experience, streamline operations, and improve service quality.
Objective: The primary goal of this project is to predict customer satisfaction using various service qualities like inflight wifi services, seat comfort, cleanliness and other flight-related variables. We aim to uncover trends and insights that can help airlines optimize their services for different customer segments. Through this project, we try to answer the below question: "How accurately can we predict the likelihood of a passenger being satisfied or dissatisfied based on demographic, flight details, and service ratings, and which specific service factors (e.g., inflight wifi, food quality, or seat comfort) have the most significant impact on these predictions?"
Dataset: The dataset used in this project has been obtained from kaggle and contains over 100,000 records of airline passengers. It includes demographic information, flight details, service ratings, and satisfaction outcomes.
Key variables include:
- Demographic Information: Gender, Age, Customer Type (loyal or disloyal).
- Flight Details: Type of Travel (business or personal), Class, Flight Distance, and delay information.
- Service Ratings: Inflight wifi, Food and drink, Seat comfort, Online boarding, and others.
- Outcome: Customer satisfaction (satisfied or neutral/dissatisfied).
Link to the dataset: https://www.kaggle.com/datasets/teejmahal20/airline-passenger-satisfaction/data
We will employ appropriate machine learning techniques to analyze and predict customer satisfaction:
- Data Preprocessing: Handle missing values, encode categorical variables, and normalize the data where necessary.
- Exploratory Data Analysis (EDA): Visualize trends, correlations, and distributions.
- Modeling: Implement models such as logistic regression, decision trees, or ensemble methods like Random Forest or XGBoost to predict customer satisfaction.
- Evaluation: Use metrics such as accuracy, precision, recall, and F1-score to evaluate model performance.