Abstract
This research presents a hybrid model for measuring reputation of organizations in online social networks (OSNs). The model combines quantitative and qualitative features that analyze OSN link structures, interactions, and users' sentiments to identify and evaluate how well an organization is perceived in a social network. The proposed reputation model is novel as it introduces several new features and a comprehensive set of metrics to calculate reputation and rank organizations. The model uses multi criteria decision making techniques to integrate the different features into one comprehensive reputation metric. The model has been implemented and applied to calculate the reputation of 47 famous worldwide companies using a dataset of more than 1.4 million posts and 7.6 million profiles collected from Twitter. The experiment shows that the proposed model outperforms other existing models and techniques as it scores the least error and the highest correlation (high accuracy) compared to a benchmark of the top 10 world's most admired companies.(c) 2022 The Authors. Published by Elsevier B.V. on behalf of King Saud University. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).