A multi-criteria decision support model for restaurant selection based on users' demand level: the case of dianping.com
The Internet, by offering a variety of information sources such as online reviews, aids people in selecting restaurants. However, it also prolongs their decision-making process due to the need to integrate information across multiple criteria. Existing decision support models for choosing satisfacto...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2024 |
| País: | España |
| Institución: | Universidad Complutense de Madrid (UCM) |
| Repositorio: | Docta Complutense |
| Idioma: | inglés |
| OAI Identifier: | oai:docta.ucm.es:20.500.14352/113741 |
| Acceso en línea: | https://hdl.handle.net/20.500.14352/113741 |
| Access Level: | acceso abierto |
| Palabra clave: | 519.8 004.6 519.226 658.8 Ordered weighted averaging aggregation operator Personalized restaurant ranking 2-tuple linguistic model Online reviews Multi-criteria decision-making Investigación Comercial Investigación operativa (Estadística) Estadística 1209.04 Teoría y Proceso de decisión 1209.03 Análisis de Datos 1209 Estadística |
| Sumario: | The Internet, by offering a variety of information sources such as online reviews, aids people in selecting restaurants. However, it also prolongs their decision-making process due to the need to integrate information across multiple criteria. Existing decision support models for choosing satisfactory restaurants overlook users' varying demand levels for each aspect of the restaurant, making the process less efficient. This paper aims to develop a multi-criteria decision support model for users to efficiently and accurately rank and select restaurants based on their demand level for various restaurant aspects. The 2-tuple linguistic ordered weighted averaging (2LOWA) aggregation operator is applied for the first time to aggregate user ratings, generating linguistic ratings that mirror the diverse levels of user demand for restaurant service, food, and environment. The importance weights (IW) method is introduced to calculate linguistic weights, thereby obtaining customized 2T-SFE composite scores under various user demand scenarios. The proposed model's applicability is demonstrated using a dataset comprising over 3.7 million reviews sourced from Dianping.com. The results show multiple personalized restaurant rankings with more linguistically understandable composite scores, enabling users to efficiently choose a suitable restaurant based on their preferences and requirements. Moreover, a list of restaurants satisfying most users with different demand levels can be generated by assessing their frequency of appearance in the top 10 restaurants across over 340 scenarios established by the proposed model. This contributes to offering more reliable and comprehensive restaurant recommendations and rankings, ultimately increasing customer satisfaction in the selection process. |
|---|