A commonsense reasoning framework for explanatory emotion attribution, generation and re-classification

We present DEGARI (Dynamic Emotion Generator And ReclassIfier), an explainable system for emotion attribution and recommendation. This relies on a recently introduced commonsense reasoning framework, the TCL logic, which is based human-like procedure automatic generation of novel concepts in Description Logics knowledge base. Starting from ontological formalization emotions Plutchik model, known as ArsEmotica, exploits logic to automatically generate semantic representations compound (e.g. Love derived combination Joy Trust according Plutchik). The generated correspond prototypes, i.e. given concepts, have been used reclassify emotion-related contents variety artistic domains, ranging art datasets editorial available RaiPlay, online platform RAI Radiotelevisione Italiana (the Italian public broadcasting company). show how reported results (evaluated light obtained reclassifications, user ratings assigned such their explainability) are encouraging, pave way many further research directions.