Analysing online reviews of restaurants in Malaysia: a novel approach to descriptive and predictive analytic

This paper aims to develop a model of restaurant products and services quality based on consumer sentiments shared on social networks. We applied term frequency-inverse document frequency (TF-IDF) weighted algorithm to generate empirical entities. These entities were incorporated into hypothetically defined constructs which reflect their thematic and sentimental nature, to test our predictive model using variance-based structural equation modelling. The results suggest that consumers have a positive attitude toward Malaysian restaurants regarding price, hospitality, location, waiting time, food variety, and restaurant atmosphere. Restaurant managers are advised to prioritise their restaurant attributes and manage key attributes to sustain and attract customers. By understanding the relative importance of restaurant reviews, restaurant managers are able to create and maintain competitive advantages in the restaurant industry, ultimately achieving customer loyalty and positive brand image.