Personality and Emotion-Aware Computing for Human-Centered AI

As humans interpose their personality into every interaction with the external world, any technology involving understanding, prediction, and synthesis of human behavior is likely to benefit from personality computing approaches, i.e., from technologies capable of dealing with human personality (Vinciarelli and Mohammadi, 2014). The recent progress of research in Artificial Intelligence (AI), and specifically in the field of Natural Language Processing (NLP) is motivating and has raised interest even from non-experts in the field (e.g., the attention ChatGPT got lately). However, the lack of in-depth analysis inspecting the actual potential of these large language systems is also concerning. For example, it is unclear to what degree they encode personality and emotion information, which leads to a lack of transparency. In this project, I will explore this potential risk, and transparently exploit this information in order to improve human-centered AI systems.