Abstract
This chapter presents an adaptive predictive model for a student modeling prediction task in the context of an adaptive educational hypermedia system (AEHS). The task, that consists in determining what kind of learning resources are more appropriate to a particular learning style, presents two issues that are critical. The first is related to the uncertainty of the information about the student's learning style acquired by psychometric instruments. The second is related to the changes over time of the student's preferences (concept drift). To approach this task, we propose a probabilistic adaptive predictive model that includes a method to handle concept drift based on statistical quality control. We claim that our approach is able to adapt quickly to changes in the student's preferences and that it should be successfully used in similar user modeling prediction tasks, where uncertainty and concept drift are presented. © 2006, Idea Group Inc.

Abstract
The availability of in effective internal similarity index to determine the "natural" number of classes in a multi spectral satellite image would benefit the process of unsupervised image classification. Two similarity indices (DB and Xu) were tested in sections of multi-spectral satellite images from Landsat TM, SPOT HRVIR, ASTER and IKONOS. The images were initially clustered into a manageable number of classes using, an unsupervised classification algorithm. These results were then structured hierarchically, and the internal similarity indices computed for each level. The inspection of the DB and Xu index plots were used to select the "natural" number of classes for each test image.

Abstract
The perception that we have of our world depends on the task we are currently performing in the environment, so if we are driving a car we will pay attention to the objects that are visually important to the task we are performing such as, the road, road signs, other vehicles, etc. The same is true when we explore virtual environments. The creation of high-fidelity 3D maps on mobile devices to aid navigation in urban environments is computationally very expensive, precluding achieving this quality at interactive rates. In this paper we present a case study to show how the human visual system may be exploited, when viewers are undertaking a task, to reduce the overall quality of the displayed image, without the users being aware of this reduction in quality. The displayed images are selectively rendered with the key features used to identify location and orientation in a 3D urban environment produced in high quality and the remainder of the image in low quality.

Abstract
This paper describes an approach towards workflow management based on the combination of learning and planning. Assuming that processes cannot be fully described at build-time, the approach makes use of learning techniques, namely Inductive Logic Programming (ILP), in order to discover workflow activities as planning operators. These operators will be subsequently fed to a partial-order planner in order to find the process model as a planning solution. The continuous interplay between learning, planning and execution aims at arriving at a feasible plan by successive refinement of the operators. The approach is illustrated in two simple scenarios. The paper concludes by relating the proposed approach with previous developments in this area.

Abstract
The aggressive application of scalar replacement to array references substantially reduces the number of memory operations at the expense of a possibly very large number of registers. In this paper we describe a register allocation algorithm that assigns registers to scalar replaced array references along the critical paths of a computation, in many cases exploiting the opportunity for concurrent memory accesses. Experimental results, for a set of image/signal processing code kernels, reveal that the proposed algorithm leads to a substantial reduction of the number of execution cycles for the corresponding hardware implementation on a contemporary Field-Programmable-Gate- Array (FPGA) when compared to other greedy allocation algorithms, in some cases, using even fewer number of registers.

Abstract
We present the classical and quantum theory of time refraction in a generic nonstationary medium. The classical approach leads to expressions for the temporal refraction coefficient, and the temporal Fresnel laws are given. The quantum formulation leads to the derivation of instantaneous Bogoliubov transformations and the evaluation of the number of photon pairs created from vacuum by the temporal changes in the medium. The influence of boundary conditions, the connection of this model with the dynamical Casimir effect, and radiation from superluminal nonaccelerated optical boundaries is also discussed.