Performance analysis using equivariant kernel density estimator in nonlinear mixture

This paper investigates the performance analysis of separation of mutually independent sources in nonlinear models. The nonlinear mapping constituted by an unsupervised linear mixture is followed by an unknown and invertible nonlinear distortion, are found in many signal processing cases. Generally, blind separation of sources from their nonlinear mixtures is rather difficult. We propose using a kernel density estimator incorporated with equivariant gradient analysis to separate the sources with nonlinear distortion. The kernel density estimator parameters of which are iteratively updated to minimize the output independence expressed as a mutual information criterion. The equivariant gradient algorithm has the form of nonlinear decorrelation to perform the convergence analysis. Experiments are proposed to illustrate these results.

How does your model grow? An empirical investigation of the effects of ontological clarity and application domain size on query performance

Semantic data models provide a map of the components of an information system. The characteristics of these models affect their usefulness for various tasks (e.g., information retrieval). The quality of information retrieval has obvious important consequences, both economic and otherwise. Traditionally, data base designers have produced parsimonious logical data models. In spite of their increased size, ontologically clearer conceptual models have been shown to facilitate better performance for both problem solving and information retrieval tasks in experimental settings. The experiments producing evidence of enhanced performance for ontologically clearer models have, however, used application domains of modest size. Data models in organizational settings are likely to be substantially larger than those used in these experiments. This research used an experiment to investigate whether the benefits of improved information retrieval performance associated with ontologically clearer models are robust as the size of the application domains increase. The experiment used an application domain of approximately twice the size as tested in prior experiments. The results indicate that, relative to the users of the parsimonious implementation, end users of the ontologically clearer implementation made significantly more semantic errors, took significantly more time to compose their queries, and were significantly less confident in the accuracy of their queries.

JKSimFloat V6.1 Plus: Improving flotation circuit performance by simulation

Analyzing, optimizing and designing flotation circuits using models and simulators have improved significantly over the last 15 years. Mineral flotation is now generally better understood through major advances in measuring and modeling the sub-processes within the flotation system. In addition, new and better methods have been derived to represent the floatability of particles as they move around a flotation circuit. A simulator has been developed that combines the effects of all of these sub-processes to predict the metallurgical performance of a flotation circuit. This paper presents an overview of the simulator, JKSimFloat V6.1PLUS, and its use in improving the industrial flotation plant performance. The application of the simulator at various operations is discussed with particular emphasis on the use of JKSimFloat V6.1PLUS in improving the flotation circuit performance.

A comparison of different choices for the regularization parameter in inverse electrocardiography models

Calculating the potentials on the heart’s epicardial surface from the body surface potentials constitutes one form of inverse problems in electrocardiography (ECG). Since these problems are ill-posed, one approach is to use zero-order Tikhonov regularization, where the squared norms of both the residual and the solution are minimized, with a relative weight determined by the regularization parameter. In this paper, we used three different methods to choose the regularization parameter in the inverse solutions of ECG. The three methods include the L-curve, the generalized cross validation (GCV) and the discrepancy principle (DP). Among them, the GCV method has received less attention in solutions to ECG inverse problems than the other methods. Since the DP approach needs knowledge of norm of noises, we used a model function to estimate the noise. The performance of various methods was compared using a concentric sphere model and a real geometry heart-torso model with a distribution of current dipoles placed inside the heart model as the source. Gaussian measurement noises were added to the body surface potentials. The results show that the three methods all produce good inverse solutions with little noise; but, as the noise increases, the DP approach produces better results than the L-curve and GCV methods, particularly in the real geometry model. Both the GCV and L-curve methods perform well in low to medium noise situations.

The Function and Consequences of Alternative Fresh Produce Supply Chain Models

Around the world, consumers and retailers of fresh produce are becoming more and more discerning about factors such as food safety and traceability, health, convenience and the sustainability of production systems, and in doing so they are changing the way in which fresh produce supply chains are configured and managed. When consumers demand fresh, safe, convenient, value-for-money produce, retailers in an increasingly competitive environment are attracted to those business models most capable of meeting these demands profitably. Traditional models are proving less and less able to deliver competitive advantage in such an environment. As a result, opportunistic, adversarial, price-based approaches to doing business between chain members are being replaced by approaches that are more strategic, collaborative and value-based. The shaping force behind this change is the need for producers, wholesalers, category managers, retailers and consumers to have more certainty about the performance of the supply chains upon which they rely. Certainty is generated through the supply chain's ability to create, deliver and share value. How to build supply chains that create, deliver and share value is arguably the single biggest challenge to the competitiveness of fresh produce firms, and therefore to the industries to which they belong.