Identifying competitors through comparative relation mining of online reviews in the restaurant industry

Abstract It is of importance for restaurants to identify their competitors to gain competitiveness. Meanwhile, opinion-rich resources like online reviews sites can be used to understand others opinion toward restaurant services. We thus propose a novel model to extract comparative relations from online reviews, and then construct three types of comparison relation networks, enabling competitiveness analysis for three tasks. The first network help restaurants analyze market structure for their positioning. The second network enables to identify top competitors using competitive index and dissimilarity index. The third network help restaurants identify strengths and weaknesses through aspects-comparison relation mining. Finally, the market environment is illustrated in a visual way according to the three types of networks. Experimental results reveal the effectiveness of the proposed competitiveness analysis using text analytics, which can identify top competitors and evaluate the market environment, as well as help the focal restaurant effectively develop a service improvement strategy.