Analysis of P2P Hosts Performance: An Integrated Deep Learning-Based Fuzzy Intuitionistic Approach

The dynamic proliferation of the peer-to-peer (P2P) hosting activities in the tourism industry has significantly attracted massive recognition by the traveling community across the globe. However, attempts to successfully evaluate P2P hosts' performance utilizing guests' online reviews (GOR) to intensify effective travel decision making has since not been externalized. As a result, this article proposes an integrated Bi-directional long short-term memory (Bi-LSTM) aggregated head attention based-interval-estimation fuzzy intuitionistic sentiment classifier to appropriately analyze P2P hosts' performance exploiting GOR. Therefore, in order to accurately yield desired output to improve subsequent processes, we initially applied time frequency-inverse document frequency and global vectorization word embedding schemes. Further, the alternatives and attributes depicting hosts performance were determined utilizing Bi-LSTM aggregated attention classifier to obtain desired polarity scores. Significantly, an interval-estimation fuzzy intuitionistic reasoning was applied to accurately enhance construction and ranking of the alternatives (hosts), major and subattributes (varied services rendered) to appropriately perform the analysis. Expressly, the conceptualization of hosting styles and segmentation of hosts was achieved using a key-seed-term level attention scheme. Ultimately, this study effectively enhances the traveling community, hosts, and the tourism sector with appropriate noesis to making informed strategic decisions.