Machine Elearning - Learning Agents and UML for Elearning Settings

We study the interplay between machine learning, agents and object-oriented design, based on the Unified Modeling Language. The application setting is electronic learning or elearning. While extending our previous object-oriented experiences, we show how these diversified technologies can be integrated and applied to elearning settings. Thus we propose to describe an early attempt of bridging the gap between web-based learning and agents capable of learning from experience. The ultimate goal sought is the development of a fully-automated multiagent environment capable of assisting in the elaboration and delivery of highly-personalized educational material effectively for anyone, anywhere at any time while taking into account each elearner's personal profile and dynamic behavior during the elearning process. We rely on software engineering paradigm to describe strategies that go from early principles to fully-developed systems. For the time being, and as far as this paper is concerned, the attempt is to concentrate on the interaction between two core fields namely Unified Modeling Language (UML) and agents. The tangible results remain the integration of agents for elearning based on machine learning methods such as Decision Tree Learning, AdaBoost, and Ensemble Learning. An emphasis is made on fuzzy agents as a special case of soft computing methods used for profile personalization. Prospectively, much effort is still required to meet the actual challenges so as to scale up to real-life problems of any significant complexity.