머신러닝과 딥러닝을 이용한 고객만족 예측모델 개발

The ICT-based 4th Industrial Revolution and the spread of COVID-19 are accelerating the big blur phenomenon in which manufacturing and service areas are gradually becoming ambiguous. In this situation, there is a need to clearly analyze data-based customer needs and expectations and reflect them in decision-making. In the industry, cases of applying actual machine learning and deep learning-based artificial intelligence to practice are emerging, but research on this is still insufficient in academia. This study is to develop a customer satisfaction prediction model using machine learning and deep learning for customer-related survey data of A coffee shop. It is also to present the process of developing a customer satisfaction prediction model. The logistic regression analysis technique used to predict when the dependent variable is bivariate (0, 1) is applied, and in this study, whether or not to visit the customer in the future (0, 1) is used as the dependent variable, and the model prediction performance is compared through comparison of learning data and verification data. Next, artificial intelligence analysis will be conducted using TensorFlow, an application program that is an open source software library produced by Google Brain Team and a library specializing in deep learning. As a result of the analysis, the accuracy of the customer satisfaction prediction model was 71.8%.