Agent-based messaging system for m-learning

Present day email encompasses a huge information space in which people spend a lot of time, and together with other message transport services (SMS, MMS, instant messaging) contribute in a significant degree to electronic information interchange. Most of the communication in the form of text or multimedia messages consists of unstructured data that is intended to humans, while the lack of semantic description prevents machine reasoning and intelligent message processing on behalf of humans. In order to extend the functionality of personal non-static e-learning systems - m-learning systems with semantic features, in this paper we propose an open and modular semantic SMS-based architecture which is based on multi-agent technology, and the requirements on the agent infrastructure for semantic SMS-based m-learning applications.