that are usually difficult to understand. Visual-ization, due to its potential to display large quantities of data, may overcome this limitation. We used the results of educational data mining techniques that had been applied to analyze the interdependence among courses in a uni-versity program and studied

State-of-the-art and emerging scientific applications require fast access to large quantities of data and commensurately fast computational resources. Both resources and data are often distributed in a wide-area network with components administered locally and independently. Computations may

Sequential quadratic programming (SQP) methods have proved highly effective for solving constrained optimization problems with smooth nonlinear functions in the objective and constraints. Here we consider problems with general inequality constraints (linear and nonlinear). We assume that first derivatives are available, and that the constraint gradients are sparse. We discuss

This paper shows that the accuracy of learned text classifiers can be improved by augmenting a small number of labeled training documents with a large pool of unlabeled documents. This is important because in many text classification problems obtaining training labels is expensive, while large

Training a support vector machine (SVM) leads to a quadratic optimization problem with bound constraints and one linear equality constraint. Despite the fact that this type of problem is well understood, there are many issues to be considered in designing an SVM learner. In particular, for large

be computed quickly, making it suitable for large data sets. Our implementation, written for use in a large-scale 3D digitization project, launches quickly, maintains a user-settable interactive frame rate regardless of object complexity or camera position, yields reasonable image quality during motion

Document clustering has not been well received as an information retrieval tool. Objections to its use fall into two main categories: first, that clustering is too slow for large corpora (with running time often quadratic in the number of documents); and second, that clustering does not appreciably

Multiscalar processors use a new, aggressive implementation paradigm for extracting large quantities of instruction level parallelism from ordinary high level language programs. A single program is divided into a collection of tasks by a combination of software and hardware. The tasks

In Western culture, we effortlessly use numerical concepts and symbols in our everyday lives – especially the positive integers. These convey a notion of preciseness and objectivity, and we readily assume that almost everything can be expressed as a numerical value. For instance,

It is now well established that, under certain conditions, the macrophage can display effector cell function toward a variety of neoplastic cells (1-5). This process is facilitated by a series of reactions that alter the functional state of the macrophage to produce an activated mononuclear phagocyte (6-8). Current evidence indicates that activation is initiated and/or augmented by a nonspecific T cell product denoted macrophage-activating factor (MAF) (9-13). 1 MAF has been shown to develop in vitro after the interaction of immune T lymphocytes with antigen-pulsed macrophages or mitogen stimulation of normal splenic cell suspensions and is thought to alter a number of functional and biochemical properties of macrophages (reviewed in 14). The biochemical identity of MAF, and therefore the elucidation of its mechanism of action on macrophage populations, have remained ill-defined. In part, this has been due to the limited amounts of MAF produced under conventional conditions, amounts insufficient to achieve a full biochemical purification of this factor. In addition, because conventional culture supernatants of activated T cells normally contain a heterogeneous mixture of lymphokines that can affect macrophage function