A_01_BAR (1) IIMBusiness analytics using R.pptx

Business analytics using R

1. Business Analytics using R Group 1 Sanskriti Gupta 2023MBA060 Jaydeep Marwah 2023MBA106 Shriya Pandey 2023MBA142 Isha Gorashiya 2023MBA261 Aarati Malve 2023MBA272 Rajashri Mondal 2023MBA286 Shalini Mondal 2023MBA291 Shivam Kumar 2023MBA294 Srija  De  2023MBA303 Varun Yadav 2023MBA312 Airline Passenger Satisfaction Prediction

2. Predicting Customer Satisfaction for an airline, based on various features related to their demographics, travel details, and service ratings. PROBLEM STATEMENT The Flight Customer Satisfaction dataset provides insights into passengers' experiences and satisfaction levels with airline services. It includes various attributes such as customer demographics, flight details, and ratings for different aspects of the travel experience. This dataset offers valuable information for airlines to understand customer preferences, improve service quality, and enhance overall satisfaction levels. Data Set Link: Airline Passenger Satisfaction Dataset

3. ANALYSIS To understand which factors are most influential we have used Logistic regression and by analysing the coefficients we have made conclusions. Also, we have used diiferent machine learning algorthims for predicting the customer satisfactions - like Random forest, Chaid, Logistic regression, C4, C4.5 etc. We have chosen a data set with 25 variables which is showcasing different factors influencing customer satisfaction like Age, Type of Travel, Travelling Class ,Flight Distance , Inflight wifi service, Departure/Arrival time convenient, Ease of Online booking, Baggage handling, Checkin service, Inflight service, Cleanliness, Departure Delay in Minutes , Arrival Delay in Minutes etc.

4. Logistic Regression Random Forest CART C4.5 C5 CHAID SVM Glmnet ALGORITHMS We have used following models in our dataset : TRAINING DATA: 80% TESTING DATA: 20% Decision Tree

5. Logistic Regression IMPLEMENTATION OF DIFFERENT ALGORITHM Glmnet

6. CHAID C4.5 IMPLEMENTATION OF DIFFERENT ALGORITHM

7. C5.0 CART IMPLEMENTATION OF DIFFERENT ALGORITHM

8. Random Forest SVM Decision Tree IMPLEMENTATION OF DIFFERENT ALGORITHM

9. Accuracy Precision Recall F-mesaure SVM 0.95109 0.96809 0.94663 0.95724 Random Forest 0.96324 0.98002 0.95608 0.96790 Logreg 0.87491 0.90436 0.87809 0.89103 GLMnet 0.87491 0.90431 0.87813 0.89103 C5.0 0.95912 0.95212 0.97683 0.96432 Cart 0.88415 0.92522 0.86545 0.89434 CHAID 0.95557 0.95163 0.97111 0.96127 COMPARISON OF DIFFERENT ALGORITHM

10. Random Forest emerges as the top-performing model with the highest accuracy, precision, recall, and F-measure. Depending on the need of project and constraints, we should select models. Random Forest model is the best choice due to its highest overall performance across all evaluation metrics. The C5.0, C4.5, and CHAID models are also highly effective. INFERENCE Arrival Delay in time , Inflight Entertainment, Online Boarding , Class these are the most influential factors in predicting satisfaction of customers.

11. Thank You