Start Date
16-8-2018 12:00 AM
Description
Online consumer reviews are becoming a key part of choosing a local business, with more consumers than ever turning to the Internet for help with everyday decisions. These reviews can help increase the visibility of the businesses, as well as provide invaluable business development insights for the owners. However, the vast amount of reviews and limited resources can make it difficult for a business to extract intelligence that helps them decide which area(s) for improvement to focus on. Previous studies have suggested that restaurant customer reviews can be categorized into multi-factors such as service quality, product quality, menu diversity, price and value, atmosphere, etc. Consequently, drawing upon eight restaurant review factors from literature and cultural restaurant reviews from a recent Yelp dataset, we propose and evaluate a content-filtering recommender system that automatically classifies individual reviews, predicts the weight and sentiment of each factor in the review, and summarizes the significant area(s) for improvement for each cultural restaurant category. We expect the findings to vary among different culture categories of restaurants. This recommender system helps to automate mining the ever growing online reviews, and provide specific business development insights for cultural restaurants. It is also potentially for other types of business with some modifications on the review factors.
Recommended Citation
Zhang, Sonya; Salehan, Mohammad; Leung, Andrew; Cabral, Ishmene; and Aghakhani, Navid, "A Recommender System for Cultural Restaurants Based on Review Factors and Review Sentiment" (2018). AMCIS 2018 Proceedings. 8.
https://aisel.aisnet.org/amcis2018/CultureIS/Presentations/8
A Recommender System for Cultural Restaurants Based on Review Factors and Review Sentiment
Online consumer reviews are becoming a key part of choosing a local business, with more consumers than ever turning to the Internet for help with everyday decisions. These reviews can help increase the visibility of the businesses, as well as provide invaluable business development insights for the owners. However, the vast amount of reviews and limited resources can make it difficult for a business to extract intelligence that helps them decide which area(s) for improvement to focus on. Previous studies have suggested that restaurant customer reviews can be categorized into multi-factors such as service quality, product quality, menu diversity, price and value, atmosphere, etc. Consequently, drawing upon eight restaurant review factors from literature and cultural restaurant reviews from a recent Yelp dataset, we propose and evaluate a content-filtering recommender system that automatically classifies individual reviews, predicts the weight and sentiment of each factor in the review, and summarizes the significant area(s) for improvement for each cultural restaurant category. We expect the findings to vary among different culture categories of restaurants. This recommender system helps to automate mining the ever growing online reviews, and provide specific business development insights for cultural restaurants. It is also potentially for other types of business with some modifications on the review factors.