The State‐of‐the‐Art in Predictive Visual Analytics

Predictive analytics embraces an extensive range of techniques including statistical modeling, machine learning, and data mining and is applied in business intelligence, public health, disaster management and response, and many other fields. To date, visualization has been broadly used to support tasks in the predictive analytics pipeline. Primary uses have been in data cleaning, exploratory analysis, and diagnostics. For example, scatterplots and bar charts are used to illustrate class distributions and responses. More recently, extensive visual analytics systems for feature selection, incremental learning, and various prediction tasks have been proposed to support the growing use of complex models, agent‐specific optimization, and comprehensive model comparison and result exploration. Such work is being driven by advances in interactive machine learning and the desire of end‐users to understand and engage with the modeling process. In this state‐of‐the‐art report, we catalogue recent advances in the visualization community for supporting predictive analytics. First, we define the scope of predictive analytics discussed in this article and describe how visual analytics can support predictive analytics tasks in a predictive visual analytics (PVA) pipeline. We then survey the literature and categorize the research with respect to the proposed PVA pipeline. Systems and techniques are evaluated in terms of their supported interactions, and interactions specific to predictive analytics are discussed. We end this report with a discussion of challenges and opportunities for future research in predictive visual analytics.     

Effect of surface exposure time on bonds to aluminum

Lap joints with AF-126 adhesive were prepared from surfaces of 2024-T3, 2024-T3 alclad, and 6061-T4 aluminum alloys treated by either FPL etch, sandblasting, or vapor degreasing. The strength data were described by a two-parameter Weibull distribution. Allowing between 1 hr and 30 days to elapse between surface preparation and actual bonding had no appreciable effect on bond strength. This was true for all three alloys surface treated in each of the three ways as well as for bonds either tested at ambient conditions or aged for 30 days at 120°F and 95% R.H. 2024-T3 aluminum, both bare and alclad, formed bonds that showed better strength than 6061-T4 aluminum.     

Preparation and characterization of poly(lactic‐co‐glycolic acid) microparticles containing DNA molecules encoding a malaria vaccine candidate

BACKGROUND: A novel ultrasonic atomization approach for the formulation of biodegradable poly(lactic‐co‐glycolic acid) (PLGA) microparticles of a malaria DNA vaccine is presented. A 40 kHz ultrasonic atomization device was used to create the microparticles from a feedstock containing 5 volumes of 0.5% w/v PLGA in acetone and 1 volume of condensed DNA which was fed at a flow rate of 18 ml h^?1. The plasmid DNA vectors encoding a malaria protein were condensed with a cationic polymer before atomization. RESULTS: High levels of gene expression in vitro were observed in COS‐7 cells transfected with condensed DNA at a nitrogen to phosphate (N/P) ratio of 10. At this N/P ratio, the condensed DNA exhibited a monodispersed nanoparticle size (Z‐average diameter of 60.8 nm) and a highly positive zeta potential of 38.8 mV. The microparticle formulations of malaria DNA vaccine were quality assessed and it was shown that the microparticles displayed high encapsulation efficiencies between 82–96% and a narrow size distribution in the range of 0.8–1.9 µm. In vitro release profile revealed that approximately 82% of the DNA was released within 30 days via a predominantly diffusion controlled mass transfer system. CONCLUSIONS: This ultrasonic atomization technique showed excellent particle size reproducibility and displayed potential as an industrially viable approach for the formulation of controlled release particles. Copyright © 2009 Society of Chemical Industry     

Preparation and assessment of frozen-hydrated sections of mammalian tissue for electron microscopy and X-ray microprobe analysis

A system is described for preparing and examining frozen-hydrated sections of mammalian tissue, using commercially available equipment. An essential part of the system is a liquid nitrogen cooled vacuum transfer system. Minor modifications were required to interface the various manufacturers' equipment, and also to facilitate handling of specimens in the cryoultramicrotome. A variety of criteria indicates that our sections are truly frozen-hydrated, but negligible image contrast in the sections is a serious problem both for morphological work and X-ray microanalysis.     

Upgrading constructed wetlands phosphorus reduction from a dairy effluent using electric arc furnace steel slag filters.

In 2003, a subsurface flow constructed wetlands (SSF-CW) system was built at the University of Vermont (UVM) Paul Miller Dairy Farm as an alternative nutrient management approach for treating barnyard runoff and milk parlour waste. Given the increasing problem of phosphorus (P) pollution in the Lake Champlain region, a slag based P-removal filter technology (PFT) was established (2004) at the CW with two objectives: (i) to test the filters' efficiency as an upgrade unit for improving P removal performance via SSF-CW (ii) to investigate the capacity of filters technology to remove P as a "stand alone" unit. Six individual filters (F1-F6) were filled with electric arc furnace (EAF) steel slag, each containing 112.5 kg of material with a pore volume of 21 L. F1-F4, fed with CW treated water, received approximately 2.17 g DRP kg(-1) EAF steel slag (0.25 kg DRP total) during the 259 day feeding period. F1-F4 retained 1.7 g DRP kg(-1) EAF steel slag, resulting in an average P removal efficiency of 75%. The addition of filters improved CW DRP removal efficiency by 74%. F5 and F6, fed non-treated water, received 1.9 g DRP kg(-1) EAF steel slag (0.22 kg DRP in total) and retained 1.5 g DRP kg(-1) resulting in a P removal efficiency of 72%. The establishment of the EAF slag based PFT is the first in-field evaluation of this technology to reduce P from dairy farm effluent in Vermont.     

An Introduction to the Community of Professors: The Engineering Education Scholars Workshop

It can be difficult for new faculty to get the information they need on issues such as teaching, advising, and setting up a research program. While some have excellent mentors, others have come to rely on trial and error or word of mouth. In 1996, the NSF Engineering Education Scholars Workshop began at Carnegie Mellon University to address the needs of new and future engineering faculty by:

• ? providing professional teaching development;
• ? offering guidance in supervising graduate students and conducting research;
• ? discussing likely engineering education and research challenges in the 21^st century; and
• ? providing intellectual and social support with colleagues.

After three years, a significant amount of knowledge and experience has been gathered by the workshop co‐chairs. This paper details the structure of the workshop and discusses the underlying principles and implementation to provide guidance for those planning similar workshops.     

New Electrocatalysts for Fuel Cells from Model Surfaces to Commercial Catalysts

In the last decade of the 20th century, we have witnessed a resurgence of interest in fuel cell technology, a technology whose basic principles were demonstrated in the mid-19th century. One of the drivers of this resurgence is the use of hydrogen as a transportation fuel, and the certification by the California Air Resources Board (CARB) of a hydrogen - air fuel cell powered vehicle as a "zero emission vehicle" or ZEV. Another driver is the simplicity of construction and assembly of fuel cell stacks when a polymer membrane is used as the electrolyte versus the more conventional aqueous electrolyte. The polymer electrolyte membrane (PEM) fuel cell offers the promise of substantially lower cost for the cell hardware. Unfortunately, two of the fundamental catalytic limitations that have plagued fuel cells for more than a century still remain: 1.) it is difficult to use any fuel other than highly purified hydrogen; 2.) there has been no cost-effective replacement of Pt as the electrocatalyst. Fortunately, there are signs that help is on the way, particularly with respect to the development of new electrocatalysts for electrooxidation of impure hydrogen, e.g. CO-contaminated, or even syngas. In this article, we describe some of these developments, which have come from over two decades of basic research on electrocatalysis.     

Biased recollections in older adults: The role of implicit theories of aging.

Explored how older adults recall the traits they possessed at an earlier age. It was hypothesized that older adults' recollections would be related to their theories about aging. In Study 1, a group of older Ss provided their theories concerning how various traits change with age. Another group of older Ss rated their current status on these traits and recalled the status they possessed at a younger age. In addition, a group of younger adults rated their current status on the same traits. On traits theorized to increase with age, older Ss recalled themselves as possessing lower levels at an earlier age than the younger group reported possessing. On traits theorized to decrease with age, older Ss recalled themselves as possessing higher levels at an earlier age than the younger group reported possessing. Study 2 indicated that this effect is obtained regardless of trait positivity. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Faster joins, self-joins and multi-way joins using join indices

We propose a new algorithm, called Stripe-join, for performing a join given a join index. Stripe-join is inspired by an algorithm called ‘Jive-join’ developed by Li and Ross. Stripe-join makes a single sequential pass through each input relation, in addition to one pass through the join index and two passes through a set of temporary files that contain tuple identifiers but no input tuples. Stripe-join performs this efficiently even when the input relations are much larger than main memory, as long as the number of blocks in main memory is of the order of the square root of the number of blocks in the participating relations. Stripe-join is particularly efficient for self-joins. To our knowledge, Stripe-join is the first algorithm that, given a join index and a relation significantly larger than main memory, can perform a self-join with just a single pass over the input relation and without storing input tuples in intermediate files. Almost all the I/O is sequential, thus minimizing the impact of seek and rotational latency. The algorithm is resistant to data skew. It can also join multiple relations while still making only a single pass over each input relation. Using a detailed cost model, Stripe-join is analyzed and compared with competing algorithms. For large input relations, Stripe-join performs significantly better than Valduriez's algorithm and hash join algorithms. We demonstrate circumstances under which Stripe-join performs significantly better than Jive-join. Unlike Jive-join, Stripe-join makes no assumptions about the order of the join index.