Research On Key Technologies Of Knowledge Grid And Its Application In Learning

The focus of the dissertation is key technologies of building knowledge grid and of assisting learning by knowledge grid. Knowledge measurement, knowledge representation, and ontology are studied. The changes in learning viewpoint and evolvement of learning system are commented on. The characteristics of learning process and information process are compared. A thorough analysis of emering technologies, their achivements, obstacles to overcome, and ongoing convergence among them, is made. Emerging technologies concerned with our purpose are grid, semantic web, service oriented architecture, semantic grid, and multi-agent system. With the purpose in mind of integrating knowledge properties, learning principles and emering technologies, a smart learning system is proposed. The reaserch is supported by the hi-tech research and development program of china under grant No. 2003AA116060. Following are contributions of the dissertation:1. A formula of computing knowledge performance is proposed. Based on the discussion about information process and learning process, a learning model is built. The performance of knowledge in the learning model is measured by the amounts of problem resolved by the knowledge. The formula proposed provides a supplement to knowledge measurement and is of importance to knowledge management, knowledge discovery, knowledge performance evaluation and knowledge learning.2. A smart learning space (SLS) based on semantic grid technology is proposed, and the Descriptive Logic model of SLS is given. With the purpose of assisting the learning activity on Internet, the SLS is composed of knowledge space, condition space and participant space. The prominent features of SLS can be claimed as (1) explicit location of learning principle, (2) technology from semantic grid, (3) foundation on Descriptive Logic, and (4) personized learning support.3. A method of providing personized learning is proposed. Learner profile space is constructed and is used as rules of selecting personized learning content. Three typical services of narrative, analogy and argument learning are proposed and designed by the method of grid compliant knowledge service (GCKS). An operation example of those services in SLS is presented.