Example-based learning

Knowledge Refinement. The first prototype is only a rough approximation of the expert s decision strategy. Many details are missing. Refinement of the prototype is accomplished by a continuation of example-based learning steps. 

A fundamentally different type of simulation is offered by science-based learning environments. Such environments incorporate some general-purpose mathematical engine that either represents nature directly or that can be programmed to represent nature. Examples are Mathematica (7) and some similar programs (8-9) for general analytical modeling in the physical sciences and Interactive Physics for introductory classical mechanics (10). Mathematica and Interactive Physics can be applied to countless topics, as opposed to the narrow focus of Flash-based simulations. Even more importantly, Mathematica and Interactive Physics are open-ended in that the software may accommodate unscripted inquiries and follow-up questions. 

In this paper, nonlinear versions of several matehed signal deteetors, sneh as KMSD, KOSP, KSMF and KASD have been implemented using the kernel-based learning theory. Performanee oomparison between the matehed signal deteetors and their eorresponding nonlinear versions was eondueted based on two-dimensional toy-examples as well as a real hyperspeetral image. It is shown that the kernel-based nonlinear versions of these deteetors outperform the linear versions. 

In instance based learning the training data is used directly and predictions are done by taking some consensus value of the nearest training set points. Sometimes called model independent methods, or naive models, they are conceptually simple and explicit examples of the well known similarity principle, which is the hypothesis that chemically similar compounds have similar properties. The diversity of methods available derives from the different choices to be made regarding how chemical space is defined and which distance metric is used. 

The measured quantities, service requests, or resource demands that are used to characterize the workload, are called workload parameters. Examples of workload parameters ate transaction types, instruction types, packet sizes, source destinations of a packet, tmd page-reference patterns. The workload parameters can be divided into workload intensity and service demands. Workload intensity is the load placed on the system, indicated by the number of units of work contending for system resources. Examples include arrival rate or interarrival times of component (e.g., transaction or request), number of clients and think times, and number of processors or threads in execution simultaneously (e.g., file reference behavior, which describes the percentage of accesses made to each file in the disk system) The service demand is the total amount of service time required by each basic component at each resource. Examples include CPU time of transaction at the database server, total transmission time of replies from the database server in LAN, and total I/O time at the Web server for requests of images and video clips used in a Web-based learning system. 

As one of the most widely available hypermedia tool kits - at least in Europe - HyperCard is becoming the prototypical environment for hypermedia courseware and hence for research in education and learning. Our aim is not to debate whether it is a proper prototype or not (for detailed analyses, see for example Jonassen, De Oliveira, this book). The issue to be discussed here deals with the relevance and effectiveness of HyperCard as a psychological research tool. Our research field is concerned with peer facilitation effects in children s individual learning. We will very briefly overview the recent trends of research in this area in order to raise some insufficiently explored issues and see how computer-based learning situations and more specifically HyperCard can help to address them. These considerations will then be illustrated by brief reports of three exploratory experiments. The discussion will finally offer some suggestions for further research in the field of Hypermedia and learning. 

---