TIU Transactions on Inteligent Computing

To dine or not to dine : Can machine learning help?

Umme Habiba Maliha, Syeda Benazir Hossain, Samsul Arefin Riffat, Sifat Momen & Shakhawat Hossain Mahi
Department of Electrical and Computer Engineering, North South University, Bangladesh


Finding a suitable restaurant to dine in can be considered as a complex decision problem for the residents of Dhaka, the capital of Bangladesh. This is due to the fact that people do not select a restaurant based on the food taste only. Rather, a range of parameters affect the decision making process. This has resulted diners to rely more on online restaurant reviews to decide on their choice. Restaurant ratings affect the customers’ decision and consequently also affect the restaurant business. As a consequence, restaurant owners are also extremely careful in maintaining the quality of service and the feedback from the customer. In this paper, we apply machine learning approach to predict the restaurant rating. Our approach finds that we can predict the rating that a customer provide with an accuracy of 92%.

Keywords: restaurant rating, classification, machine learning, prediction