In this paper we discuss the integration of a communication model in the MOBOT assistive robotic platform and its evaluation by target users. The MOBOT platform envisions the development of cognitive robotic assistant prototypes that act proactively, adaptively and interactively with respect to elderly humans with slight walking and cognitive impairments. The respective multimodal action recognition system has been developed to monitor, analyze and predict user actions with a high level of accuracy and detail. The robotic platform incorporates a human-robot communication model that has been defined with semantics of human actions in interaction, their capture and their representation in terms of behavioral patterns, to achieve an effective, natural interaction, aiming to support elderly users of slight walking and cognitive inability. The platform has been evaluated in a series of validation experiments with end users, the procedure and results of which are also presented in this paper.
Detecting emotional traits in call centre interactions can be beneficial to the quality management of the services provided, since this reveals the positioning of both speakers, i.e. satisfaction or frustration and anger on the customers' side, and stress detection, disappointment mitigation or failure to provide the requested service on the operators' side. This paper describes a machine learning approach to classify emotional speech units occurring in a call centre dataset by employing emotion-related labels, automatically extracted acoustic features as well as additional context-related features.
In this study, we define and test measures that capture aspects of collaboration in interaction within groups of three participants performing a task. The measures are constructed upon turn-taking and lexical features from a corpus of triadic task-based interactions, as well as upon demographic features, and personality, dominance, and satisfaction assessments related to the corpus participants. Those quantities were tested for significant effects and correlations they have with each other. The findings indicate that determinants of collaboration are located in measures quantifying the differences among dialogue participants in conversational mechanisms employed, such as number and frequency of words contributed, lexical repetitions, conversational dominance, and in psychological and sentiment variables, i.e., the participants’ personality traits and expression of satisfaction.