A picture fuzzy set multi criteria decision-making approach to customize hospital recommendations based on patient feedback

Sentiment analysis techniques have allowed exploiting the information available in millions of opinions conveyed through different Internet services. One example would be the multiple opinions about medical experiences in hospitals available on the website called Careopinion. These opinions usually talk about different medical aspects such as staff, facilities, etc., in a positive, negative, or neutral manner. Nevertheless, there are situations in which the same opinion contains positive, neutral and negative ideas regarding the same aspect. This fact leads to a perception of hesitancy and uncertainty about the opinion. To deal with this issue, this study proposes a picture fuzzy set-based model able to represent this hesitancy in terms of polarity values. To test this model, it has been used to implement a multicriteria decision making-based hospital recommender which considers the patient preferences with respect to the aspects of the hospitals. The proposed approach has been tested using real reviews from 8 hospitals considering diverse patient preferences. The results of all experiments were compared against an ideal ranking computed from the patient reviews using Spearman's footrule. Furthermore, to assess the effectiveness of the proposal, it has been compared against other state-ofthe-art logic-based polarity representation mechanisms. The findings demonstrate that the proposed approach is more effective than the other polarity representation methods by at least 4%, confirming the superiority of the proposed approach to capture and represent sentiments in an accurate manner.