An ontology-based affective computing approach for passenger safety engagement on cruise ships

The safety of cruise passengers is a key element for a cruise company. Among various aspects, the ability to interpret and recognize cruisers' emotions, so that he/she feel safe, plays a central role in human communication. Affective computing addresses the computational processing of emotions. Current automatic emotion recognizers basically use automated classification tools to label emotions without capturing relations between biosignals and observations measured by the various sensors. This work proposes instead an Ontology Web Language (OWL) ontology-based emotion recognition framework by (i) monitoring human body vital signals through wearable, noninvasive sensors; and then the (ii) emotion detection is based on a ontology-based matchmaking process via non-standard automated reasoning services. A key factor is the use of so-called vague/fuzzy concepts, which are intrinsic in the realm of emotions and their dynamic evolution. To this end, we exploit Fuzzy Description Logics (Fuzzy DLs), which are the logical foundation of fuzzy OWL ontologies, i.e., OWL ontologies extended with vague/fuzzy concepts. An early prototype has been implemented w.r.t. a reference dataset and a preliminary experiment has being carried out with the aim to monitor the emotions experienced by cruise passengers while viewing safety video instructions.