Authors: | Papakyriakopoulos, Dimitrios Novas, Dimitrios Griva, Anastasia Kartaloglou, Elissavet |
Issue Date: | 1-Jan-2023 |
Journal: | International Journal of Multicriteria Decision Making |
Volume: | 9 |
Issue: | 3 |
Keywords: | Fuzzy numbers, Multi-criteria analysis, Ranking, Restaurants, Text mining, TripAdvisor |
Abstract: | User generated content (UGC) is a valuable resource for the multi-criteria decision-making process. Platforms, such as TripAdvisor, enable the registration of comments and post rankings. However, existing ranking mechanisms are not transparent, making it impossible to evaluate their integrity and accuracy. Additionally, it has been shown that these rankings are, in some cases, inaccurate and misleading. This prompted the need for developing a ranking mechanism based on UGC to capture experiences and feelings, rather than quantitative data (bubbles/stars). The proposed ranking mechanism employs LDA to generate topics, which are transformed with fuzzy logic to variables, used to produce ranking results. We empirically studied the proposed ranking mechanism using TripAdvisor's user comments on a sample of restaurants located in Athens, Greece and compared the results to a simple quantitative ranking scheme. In some cases, the ranking differed. Further investigation is needed to address the limitations encountered in this research. |
ISSN: | 2040-1078 2040-106X |
DOI: | 10.1504/IJMCDM.2023.134135 |
URI: | https://uniwacris.uniwa.gr/handle/3000/2148 |
Type: | Article |
Department: | Department of Business Administration |
School: | School of Administrative, Economics and Social Sciences |
Affiliation: | University of West Attica (UNIWA) |
Appears in Collections: | Articles / Άρθρα |
CORE Recommender
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.