A dynamic group decision making process for high number of alternatives using hesitant Fuzzy Ontologies and sentiment analysis

• Published in: Knowledge-based systems 2020-05, Vol.195, p.105657, Article 105657
• Main Authors: Morente-Molinera, J.A., Cabrerizo, F.J., Mezei, J., Carlsson, C., Herrera-Viedma, E.
• Format: Article
• Language: English
• Subjects: Computing with words; Data mining; Decision making; Fuzzy ontologies; Fuzzy sets; Group decision making; Knowledge representation; Ontology; Sentiment analysis; Social networks; Type-2 hesitant fuzzy sets
• ISSN: 0950-7051; 1872-7409
• DOI: 10.1016/j.knosys.2020.105657

The high spread of Internet and social networks have completely changed the way that Group Decision Making methods are designed, developed and implemented. Experts now operate in environments where a large amount of information is available and new ideas and participants can appear at any time; this results in a dynamically changing decision environment. In this paper, a novel group decision making method for dynamic contexts with a high number of decision alternatives is presented. As the main component of the proposal, a perceptual computing scheme is used in order to extract information from the experts. In the process, sentiment analysis is used when analyzing the debate texts in order to obtain information for selecting the best alternatives on each round. Moreover, interval type-2 hesitant Fuzzy Ontologies are used in order to store the information related to alternatives. By combining interval type-2 and hesitant fuzzy sets, imprecise information can be represented in a comfortable and intuitive way within the ontology. •A novel Group Decision Making model for high number of alternatives.•Sentiment analysis is used for extracting information directly from debate.•A novel fuzzy ontology model is presented.•Interval type-2 fuzzy sets are used for increasing flexibility in representation.