Restaurant rating prediction through features : a Zomato case

Abstract

This research attempts to examine the impact of restaurants features on their online rating in order to build a rating prediction model. We extracted our data from Zomatos website and focused on Beiruts restaurants. The nature of our research is quantitative with a cross sectional research design. We processed the data collected through multiple regression to obtain the optimum model with the most influential features. In our findings, popularity had the most presence with a positive effect on the overall, service and look and feel ratings. Interestingly enough, the price index had an impact only on the service rating with no effect on other ratings. Another interesting find was that the food related services index had no significance in relation to the restaurants ratings. Finally, from a theoretical point, the SERVQUAL model was utilized only as a dimensions explanatory aid without the intent to accept or reject the theory. On the other hand the practical implications of our research would be considered by restaurants owners, consultants and entrepreneurs as a guide to earn the highest ratings.