The use of Multi Agent Systems in Cloud e-Learning

In recent years, personalization and customization in cloud services are more than evident. Cloud e-Learning (CeL) is an upgraded model of traditional e-Learning which aims to offer personalization, collaboration through interaction and structured customized learning content. Cloud e- Learners will express their preferences and be offered courses as a result of a multi-agent system (MAS) interaction which will match learners’ profiles and preferences with content of the learning cloud. The level of intelligence of the multi-agent system will continually increase as more learners interact with the system. Actual learning will be evaluated by agents and learners, thus providing opportunities to change preferences, domain interests and desired level of knowledge. Agents will be semantically assisted by ontologies. In this paper, we define a scenario for CeL and demonstrate the need for MAS and semantic annotation. We also present a tentative abstract architecture and speculate on future work.

近年来，云服务领域的个性化与定制化趋势已愈发显著。云电子学习（Cloud e-Learning，CeL）作为传统电子学习的升级模式，旨在提供个性化服务、基于交互的协作体验以及结构化定制化学习内容。云学习学习者可表达自身偏好，多智能体系统（Multi-agent System，MAS）会通过交互将学习者的画像与偏好与学习云平台的内容进行匹配，进而为学习者推送适配课程。随着更多学习者与该系统交互，多智能体系统的智能水平将持续提升。实际学习效果将由智能体与学习者共同评估，从而为调整偏好、领域兴趣以及预期知识水平提供契机。智能体将依托本体（Ontology）实现语义层面的辅助支持。本文针对云电子学习构建了应用场景，并论证了多智能体系统与语义标注的必要性。此外，本文还提出了一套初步的抽象架构，并对未来研究方向进行了展望。