Parametric embedding for class visualization

We propose a new method, parametric embedding (PE), that embeds objects with the class structure into a low-dimensional visualization space. PE takes as input a set of class conditional probabilities for given data points and tries to preserve the structure in an embedding space by minimizing a sum of Kullback-Leibler divergences, under the assumption that samples are generated by a gaussian mixture with equal covariances in the embedding space. PE has many potential uses depending on the source of the input data, providing insight into the classifier's behavior in supervised, semisupervised, and unsupervised settings. The PE algorithm has a computational advantage over conventional embedding methods based on pairwise object relations since its complexity scales with the product of the number of objects and the number of classes. We demonstrate PE by visualizing supervised categorization of Web pages, semisupervised categorization of digits, and the relations of words and latent topics found by an unsupervised algorithm, latent Dirichlet allocation.     

Learning and retention of associations between auditory icons and denotative referents: implications for the design of auditory warnings

OBJECTIVE: This study examined the way in which the type and preexisting strength of association between an auditory icon and a warning event affects the ease with which the icon/event pairing can be learned and retained. BACKGROUND: To be effective, an auditory warning must be audible, identifiable, interpretable, and heeded. Warnings consisting of familiar environmental sounds, or auditory icons, have potential to facilitate identification and interpretation. The ease with which pairings between auditory icons and warning events can be learned and retained is likely to depend on the type and strength of the preexisting icon/event association. METHOD: Sixty-three participants each learned eight auditory-icon/denotative-referent pairings and attempted to recall them 4 weeks later. Three icon/denotative-referent association types (direct, related, and unrelated) were employed. Participants rated the strength of the association for each pairing on a 7-point scale. RESULTS: The number of errors made while learning pairings was greater for unrelated than for either related or direct associations, whereas the number of errors made while attempting to recall pairings 4 weeks later was greater for unrelated than for related associations and for related than for direct associations. Irrespective of association type, both learning and retention performance remained at very high levels, provided the strength of the association was rated greater than 5. CONCLUSION: This suggests that strong preexisting associations are used to facilitate learning and retention of icon/denotative-referent pairings. APPLICATION: The practical implication of this study is that auditory icons having either direct or strong, indirect associations with warning events should be preferred.     

Connection and coordination: the interplay between architecture and dynamics in evolved model pattern generators

We undertake a systematic study of the role of neural architecture in shaping the dynamics of evolved model pattern generators for a walking task. First, we consider the minimum number of connections necessary to achieve high performance on this task. Next, we identify architectural motifs associated with high fitness. We then examine how high-fitness architectures differ in their ability to evolve. Finally, we demonstrate the existence of distinct parameter subgroups in some architectures and show that these subgroups are characterized by differences in neuron excitabilities and connection signs.     

The mechanical properties and tensile failure mechanism of a high strength polymer modified Portland cement

The mechanical properties of one of the new high strength polymer modified hydraulic cements have been investigated. An important parameter for the material is the amount of polymer present, and the properties are found to be dependent upon the degree of drying. For example, in the wet state, polymer content has little effect upon ultimate flexural strength, but does cause nonlinearity in the stress-strain behaviour. Although increasing polymer content causes a reduction in the initial tangent modulus, it is shown that retardation of hydration may account for this. In the dry state, increasing polymer content leads to increasing flexural strength, fracture toughness and failure strain, but leaves initial elastic modulus relatively unchanged. It is concluded that removal of pores is not the principal strengthening mechanism since strength increases are consequent upon water removal from the microstructure in the presence of the polymer. A fibrillar pull-out model is proposed to explain the observed behaviour of both wet and dry material and ordinary Portland cement paste, which shows good correlation with experimental results. The principal effect of the polymer is to act as an adhesive at the interface between interacting CSH fibrils.     

Lysosomal proteinases in muscle tissue and leukocytes of meat animals

Carboxypeptidases A and B, cathepsins B, D and E, dipeptidyl aminopeptidase 1 and collagenase were extracted from bovine and porcine leukocyte lysosomes and diaphragm muscle lysosomes with 4% n-butyl alcohol in dry ice and various chemical properties were studied. The enzymes were classified by their specific activities with different substrates, pH optima, energies of activation and the influence of inhibitors and stimulators. Cathepsins D and E were the most active proteases studied and all enzymes were more active in leukocytes than in sediments from muscle tissue. The properties of these enzymes, compared with enzymes from other sources, and their possible involvement in meat tenderness are discussed.     

Spatial and Temporal Variabilities of Sediment Delivery Ratio

Sediment contribution of specific areas of a watershed is an important consideration, but one that is often overlooked, when developing watershed management plans. Understanding sediment delivery to the watershed outlet is one method for identifying high contribution areas. In this study, a new methodology for calculating individual subbasin sediment delivery ratio (SDR) was developed within the Soil and Water Assessment Tool (SWAT) for two Michigan watersheds. Subbasins with the greatest SDR had two defining characteristics: high erosion rates and close proximity to the watershed outlet. The impact of best management practice (BMP) implementation (no-tillage and Conservation Reserve Program) on sediment yield reduction at the watershed outlet was compared for two targeting methods (SDR and erosion rate). Identifying SDR of individual subbasins for implementation of BMPs is more effective method for addressing water quality concerns at the watershed outlet than the widely used targeting high risk erosion areas. Watershed SDR was found to be highly variable on a monthly basis, due to changes in sediment yield, which is in turn affected by factors such as precipitation and surface runoff. Finally, watershed SDR calculated using the physically-based SWAT model was compared to four empirically-based areal relationships with SDR. While some area methods reproduced the SWAT-SDR, the variations between areal methods indicate that more detailed physical characteristics should be considered in watershed SDR calculation.     

Prediction of backside micromotion in total knee replacements by finite element simulation

The micromotion at the interface between the polyethylene tibial insert and metal tibial tray [corrected] in modular total knee replacements [corrected] has been shown to contribute to wear particle-induced osteolysis and may [corrected] cause implant failure. Therefore, studying the design parameters that are involved in the backside wear process is an important task that may lead to improvement in new total knee replacements. In the present study, a finite element model was developed to predict the backside micromotion along the entire modular interface. Both the linear elastic constitutive model and non-linear J2-plasticity constitutive model were considered in the finite element model for polyethylene and were corroborated against published results obtained from displacement controlled knee simulator wear tests. The finite element simulation with the non-linear J2-plasticity constitutive model was able to predict backside micromotion [corrected] more accurately than the simulation with the linear elastic constitutive model. [corrected] The developed finite element model (including the non-linear J2-plasticity constitutive model) was then applied to assess the effects of the tibial tray locking mechanism design (dovetails versus fullperipheral [corrected] design) and different levels of interference fit on insert micromotion. The developed finite element model, implementing the non-linear J2-plasticity constitutive model, was shown to successfully predict clinical amounts of backside micromotion and could be used for the design and development of total knee replacements for the reduction of backside micromotion and polyethylene [corrected] wear.     

Examining cognitive function across the lifespan using a mobile application

Many studies conducted in a laboratory or university setting are limited by funding, personnel, space, and time constraints. In the present study, we introduce a method of data collection using a mobile application that circumvents these typical experiment administration issues. Using the application, we examined cross-sectional age differences in cognitive function. We obtained data from more than 15,000 participants and replicated specific patterns of age-related differences in cognition. Using a subset of these participants, we also examined the processing speed account of age-related cognitive differences, and the association of exercise and leisure activity with cognitive function across the lifespan. We discuss the relative advantages and disadvantages of data collection with a mobile application, and provide recommendations for the use of this method in research.     

Exploring the impact of software requirements on system-wide goals: a method using satisfaction arguments and i* goal modelling

This paper describes the application of requirements engineering concepts to support the analysis of the impact of new software systems on system-wide goals. Requirements on a new or revised software component of a socio-technical system not only have implications on the goals of the subsystem itself, but they also impact upon the goals of the existing integrated system. In industries such as air traffic management and healthcare, impacts need to be identified and demonstrated in order to assess concerns such as risk, safety, and accuracy. A method called PiLGRIM was developed which integrates means-end relationships within goal modelling with knowledge associated with the application domain. The relationship between domain knowledge and requirements, as described in a satisfaction argument, adds traceability rationale to help determine the impacts of new requirements across a network of heterogeneous actors. We report procedures that human analysts follow to use the concepts of satisfaction arguments in a software tool for i* goal modelling. Results were demonstrated using models and arguments developed in two case studies, each featuring a distinct socio-technical system??a new controlled airspace infringement detection tool for NATS (the UK??s air navigation service provider), and a new version of the UK??s HIV/AIDS patient reporting system. Results provided evidence towards our claims that the conceptual integration of i* and satisfaction arguments is usable and useful to human analysts, and that the PiLGRIM impact analysis procedures and tool support are effective and scalable to model and analyse large and complex socio-technical systems.