Predicting Customer Satisfaction based on Neutrosophic sets:  Applied on mobile food ordering application

Walaa Elsanabary; Nouran M. Radwan

Authors

Walaa Elsanabary Computer and Information Systems Dept.; Sadat Academy for Management Sciences;
Nouran M. Radwan Computer and Information Systems Dept.; Sadat Academy for Management Sciences;

Keywords:

Food Ordering Applications, Neutrosophic Logic, Neutrosophic set, Classification, Neutrosophic Classifier.

Abstract

Customers’ satisfaction prediction is a vital process for all business organizations to draw
new customers and maintain existing customers. One of the most efficient methods to predict
customers' satisfaction is classifying customers' feedback. In the real world, customer’s feedback is
ambiguous, confusing and inconsistent. This option can be stated in neutrosophic logic as
indeterminacy membership, associated with truth and falsity membership. In this study, a
classification model based on neutrosophic sets to handle the inconsistency of customer responses
is presented. Also, significant factors that impact customers' satisfaction are defined. In order to
show the procedures and the application of the proposed method, a case study to determine the
customers' satisfaction while using food order application (Talabat) in Egypt is presented. A
comparison between the classical classification models and the proposed model based on
neutrosophic sets is presented. The experimental results indicate that the proposed classifying
model achieved accuracy results around 95.36% to 99.95%, higher than the classical one that achieve
around 90.3% to 93.99%. Next, a sensitivity analysis is performed for reliability validation as to
determine the most factors that affect customers’ satisfaction.

DOI: 10.5281/zenodo.14707541