Design and Evaluation of a Conversational Assistant for Supporting Erasmus+ Student-Exchange Volunteers
Publication Type:
Conference PaperSource:
11th South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference, IEEE, Thessaloniki, Greece (2026)Keywords:
Chatbots, Cognitive workload, Conversational agents, Erasmus+ mobility, Student support, Usability evaluationAbstract:
At the University of Patras, Greece, incoming Erasmus+ students for studies rose by about 172 percent between 2014 and 2025 but the number of local Erasmus+ Student Network (ESN) volunteers that offer crucial assistance and support for incoming students, has remained nearly identical. Local volunteers often cannot answer routine questions whilst, the relevant information is scattered across documents and websites that students rarely consult. In response to this obstacle, this work presents the design and evaluation of a domain-specific conversational assistant, built with the open-source Rasa framework on a documented analysis of real questions, that provides immediate vetted answers. Its aim is not to replace local volunteers, but instead support their work in helping incoming students. The evaluation combined a subjective study of user experience with thirty-four participants and a counterbalanced within-subjects experiment with twenty volunteers that compared answering tasks with and without the assistant using the NASA Task Load Index and objective interaction logs. Participants rated the experience highly with the weighted workload dropping significantly. The reduction was larger for more experienced ESN members. The assistant works as a complement to human support which can help to reduce the effort and frustration of question answering.
