Restaurant recommendation model using textual information to estimate consumer preference: evidence from an online restaurant platform

• DOI: 10.1108/jhtt-01-2023-0019
• Publication Date: 2023-08-01T04:40:45Z
• Authors: Qinglong Li; Dongsoo Jang; Dongeon Kim; Jaekyeong Kim
• Keywords: Consumer behaviour

Purpose Textual information about restaurants, such as online reviews and food categories, is essential for consumer purchase decisions. However, previous restaurant recommendation studies have failed to use textual containing predicting preferences effectively. This study aims propose a novel model effectively estimate the assessment behaviors of consumers multiple attributes. Design/methodology/approach The authors collected 1,206,587 from 25,369 46,613 restaurants Yelp.com. Using these data, generated preference vector by combining identity reviews. Thereafter, combined categories generate vector. Finally, nonlinear interaction between vectors was learned considering attribute Findings found that proposed exhibited excellent performance compared with state-of-the-art models, suggesting various on fundamental factor in determining predictions. Originality/value To best authors’ knowledge, this first develop personalized using real-world platforms. also presents deep learning mechanisms outperform models. results can reduce cost exploring support effective purchasing