OpenTable data with multi-criteria ratings
Yong Zheng
TL;DR
The paper addresses the need for real-world benchmarks for multi-criteria recommender systems by releasing the OpenTable dataset. It details data collection via web crawling and includes an overall rating plus four per-criterion ratings for 19,536 ratings across 1,309 users and 91 restaurants, with high sparsity. The dataset can support traditional RS using overall ratings and multi-criteria RS using both overall and per-criterion ratings, with prior evaluations conducted via the MCRecKit library. It also discusses data cleaning challenges from anonymous web sources and provides a cleaned opentable_cleaned.csv to ensure reproducibility. Overall, the work provides a concrete real-world benchmark for evaluating MCRSs in the restaurant domain and facilitates method comparison and reproducibility.
Abstract
With the development of recommender systems (RSs), several promising systems have emerged, such as context-aware RS, multi-criteria RS, and group RS. Multi-criteria recommender systems (MCRSs) are designed to provide personalized recommendations by considering user preferences in multiple attributes or criteria simultaneously. Unlike traditional RSs that typically focus on a single rating, these systems help users make more informed decisions by considering their diverse preferences and needs across various dimensions. In this article, we release the OpenTable data set which was crawled from OpenTable.com. The data set can be considered as a benchmark data set for multi-criteria recommendations.