Framework for cognitive agent based expert system for metacognitive and collaborative E-Learning

Abstract

Globalization, Customization and Web 2.01 are having substantial effects in the Education sector. For this, the government of India has taken strong steps towards Collaborative E-Learning which yields significant social, educational, financial and research profits. The learning infrastructure has improved many folds providing easy access to innovative technologies. But, for its utmost utilization, there is need to identify the intended learners and their cognitive requirements to produce precise E-Content and Content Management Systems (CMS)2 to match the learner's needs, cognitive ability and learning attitude. Keeping this in view, the present study suggests a cognitive agent3 based expert system for collaborative E-Learning. The system drives on learner's cognitive ability. It simulates the human thought process during teaching-learning activity and collaborative learning; and thus, provides the relevant, unambiguous E-Content. The system can be used to obtain an insight to the learner's cognitive aspects and willingness to learn in social interactive environment. Based on the empirical data gathered, the system instructs the learner for self paced learning or social interactive learning. The system is referred as CAESMCE-Learning (Cognitive Agent based Expert System for Metacognitive and Collaborative E-Learning). The agent acts as a software artifact that exhibits intelligent behavior in a very complex domain like human learning where the system attempts to model the cognitive process associated with learning. The system is like an intelligent agent which is reactive, instructable, adaptive and cognitive. The property of instructability is important since by allowing the instructional designer to feed the agent with new knowledge during its execution, it helps to solve the knowledge acquisition problem. The system models human intelligence and human perspective of the world using BDI (Belief-Desire-Intention) Model. In the context of this study, Metacognitive E-Learning is a computer-supported self paced collaborative learning approach which assists the learner to become more aware of the learning attitude and learning requirements and thus, make strategies for learning based on the experience. Metacognitive approach can help the learners to help themselves. This system can be used to determine the learning pattern and thought process of the learner during the learning activity. It can be used to study the impact of self paced and collaborative E-Learning on learner's cognitive capability. It is no longer confined to the delivery of instructional packets delivered to learners and then evaluated by the teacher. Now, the emphasis is on self paced social learning using social software such as blogs, wikis, podcasts. It works on the principle that the knowledge is socially constructed and learning takes place through conversations about content and grounded interaction about problems and actions. It is believed that one of the best ways to learn something is to teach it to others. Cognitive agents are best suited to solve such complicated problems in the domain that requires extensive human expertise and time. They simulate the human reasoning process with the help of knowledge base and inference engine. Education system would be revolutionized with the use of Cognitive agent based systems for learning, teaching and planning. Metacognition refers to the knowledge concerning one's own cognitive processes and products or anything related to them [5].