A case is made for the importance of high quality semantic and coreference annotation. The challenges of providing such annotation are described. Asperger's Syndrome is introduced, and the connections are drawn between the needs of text annotation and the abilities of persons with Asperger's Syndrome to meet those needs. Finally, a pilot program is recommended wherein semantic annotation is performed by people with Asperger's Syndrome. The primary points embodied in this paper are as follows: (1) Document annotation is essential to the Natural Language Processing (NLP) projects at Lawrence Livermore National Laboratory (LLNL); (2) LLNL does not currently have a system in place to meet its need for text annotation; (3) Text annotation is challenging for a variety of reasons, many related to its very rote nature; (4) Persons with Asperger's Syndrome are particularly skilled at rote verbal tasks, and behavioral experts agree that they would excel at text annotation; and (6) A pilot study is recommend in which two to three people with Asperger's Syndrome annotate documents and then the quality and throughput of their work is evaluated relative to that of their neuro-typical peers.

This paper summarizes an exercise carried out to validate the process of implementing LLNL's ALE3D to predict the permanent deformation and rupture of an air-backed steel plate subjected to underwater shock. Experiments were performed in a shock tank at the Naval Science and Technology Laboratory in Visakhapatnam India, and the results are documented in reference. A consistent set of air-backed plates is subjected to shocks from increasing weights of explosives ranging from 10g-80g. At 40g and above, rupture is recorded in the experiment and, without fracture mechanics implemented in ALE3D, only the cases of 10g, 20g, and 30g are presented here. This methodology applies the Jones-Wilkins-Lee (JWL) Equation of State (EOS) to predict the pressure of the expanding detonation products, the Gruneisein EOS for water under highly dynamic compressible flow - both on 1-point integrated 3-d continuum elements. The steel plates apply a bilinear elastic-plastic response with failure and are simulated with 3-point integrated shell elements. The failure for this exercise is based on effective (or equivalent) plastic strain.

Probabilistic inverse techniques, like the Markov Chain Monte Carlo (MCMC) algorithm, have had recent success in combining disparate data types into a consistent model. The Stochastic Engine (SE) initiative was a technique that developed this method and applied it to a number of earth science and national security applications. For instance, while the method was originally developed to solve ground flow problems (Aines et al.), it has also been applied to atmospheric modeling and engineering problems. The investigators of this proposal have applied the SE to regional-scale lithospheric earth models, which have applications to hazard analysis and nuclear explosion monitoring. While this broad applicability is appealing, tailoring the method for each application is inefficient and time-consuming. Stochastic methods invert data by probabilistically sampling the model space and comparing observations predicted by the proposed model to observed data and preferentially accepting models that produce a good fit, generating a posterior distribution. In other words, the method ''inverts'' for a model or, more precisely, a distribution of models, by a series of forward calculations. While powerful, the technique is often challenging to implement, as the mapping from model space to data needs to be ''customized'' for each data type. For example, all …

The main objective of this project was to evaluate pretargeting systems that use a bispecific antibody (bsMAb) to improve the detection and treatment of cancer. A bsMAb has specificity to a tumor antigen, which is used to bind the tumor, while the other specificity is to a peptide that can be radiolabeled. Pretargeting is the process by which the unlabeled bsMAb is given first, and after a sufficient time (1-2 days) is given for it to localize in the tumor and clear from the blood, a small molecular weight radiolabeled peptide is given. According to a dynamic imaging study using a 99mTc-labeled peptide, the radiolabeled peptide localizes in the tumor in less than 1 hour, with > 80% of it clearing from the blood and body within this same time. Tumor/nontumor targeting ratios that are nearly 50 times better than that with a directly radiolabeled Fab fragment have been observed (Sharkey et al., ''Signal amplification in molecular imaging by a multivalent bispecific nanobody'' submitted). The bsMAbs used in this project have been composed of 3 antibodies that will target antigens found in colorectal and pancreatic cancers (CEA, CSAp, and MUC1). For the ''peptide binding moiety'' of the bsMAb, we initially …

The most important requirement for high-level waste glass acceptance for disposal in a geological repository is the chemical durability, expressed as a glass dissolution rate. During the early stages of glass dissolution in near static conditions that represent a repository disposal environment, a gel layer resembling a membrane forms on the glass surface through which ions exchange between the glass and the leachant. The hydrated gel layer exhibits acid/base properties which are manifested as the pH dependence of the thickness and nature of the gel layer. The gel layer has been found to age into either clay mineral assemblages or zeolite mineral assemblages. The formation of one phase preferentially over the other has been experimentally related to changes in the pH of the leachant and related to the relative amounts of Al{sup +3} and Fe{sup +3} in a glass. The formation of clay mineral assemblages on the leached glass surface layers ,lower pH and Fe{sup +3} rich glasses, causes the dissolution rate to slow to a long-term steady state rate. The formation of zeolite mineral assemblages ,higher pH and Al{sup +3} rich glasses, on leached glass surface layers causes the dissolution rate to increase and return to the initial high …

Michigan Technological University, together with The Robbins Group, Advanced Ceramic Research, Advanced Ceramic Manufacturing, and Superior Rock Bits, evaluated a new process and a new material for producing drill bit inserts and disc cutters for the mining industry. Difficulties in the material preparation stage slowed the research initially. Prototype testing of the drill bit inserts showed that the new inserts did not perform up to the current state of the art. Due to difficulties in the prototype production of the disc cutters, the disc cutter was manufactured but not tested. Although much promising information was obtained as a result of this project, the objective of developing an effective means for producing rock drill bits and rock disc cutters that last longer, increase energy efficiency and penetration rate, and lower overall production cost was not met.

We describe a new approach to scalable data analysis that enables scientists to manage the explosion in size and complexity of scientific data produced by experiments and simulations. Our approach uses a novel combination of efficient query technology and visualization infrastructure. The combination of bit map indexing, which is a data management technology that accelerates queries on large scientific datasets, with a visualization pipeline for generating images of abstract data results in a tool suitable for use by scientists in fields where data size and complexity poses a barrier to efficient analysis. Our architecture and implementation, which we call DEX (short for dexterous data explorer), directly addresses the problem of ''too much data'' by focusing analysis on data deemed to be ''scientifically interesting'' via a user-specified selection criteria. The architectural concepts and implementation are applicable to wide variety of scientific data analysis and visualization applications. This paper presents an architectural overview of the system along with an analysis showing substantial performance over traditional visualization pipelines. While performance gains are a significant result, even more important is the new functionality not present in any visualization analysis software--namely the ability to perform interactive, multi-dimensional queries to refine regions of interest that are …

Numerical modeling and field data tests are presented on the Transfer Function/Scattering Index Method for estimating fracture orientation and density in subsurface reservoirs from the ''coda'' or scattered energy in the seismic trace. Azimuthal stacks indicate that scattered energy is enhanced along the fracture strike direction. A transfer function method is used to more effectively indicate fracture orientation. The transfer function method, which involves a comparison of the seismic signature above and below a reservoir interval, effectively eliminates overburden effects and acquisition imprints in the analysis. The transfer function signature is simplified into a scattering index attribute value that gives fracture orientation and spatial variations of the fracture density within a field. The method is applied to two field data sets, a 3-D Ocean Bottom Cable (OBC) seismic data set from an offshore fractured carbonate reservoir in the Adriatic Sea and a 3-D seismic data set from an onshore fractured carbonate field in the Middle East. Scattering index values are computed in both fields at the reservoir level, and the results are compared to borehole breakout data and Formation MicroImager (FMI) logs in nearby wells. In both cases the scattering index results are in very good agreement with the well …

When metallic plates are welded, residual tensile stresses may develop in the vicinity of the weld seam. Processes such as Low Plasticity Burnishing (LPB) and Laser Shock Peening (LSP) could be applied locally to eliminate the residual stresses produced by welding. In this study, Alloy 22 (N06022) plates were welded and then the above-mentioned surface treatments were applied to eliminate the residual tensile stresses. The aim of the current study was to compare the corrosion behavior of as-welded (ASW) plates with the corrosion behavior of plates with stress mitigated surfaces. Immersion and electrochemical tests were performed. Results show that the corrosion resistance of the mitigated plates was not affected by the surface treatments applied.

The field of Natural Language Processing (NLP) is described as it applies to the needs of LLNL in handling free-text. The state of the practice is outlined with the emphasis placed on two specific aspects of NLP: Information Extraction and Discourse Integration. A brief description is included of the NLP applications currently being used at LLNL. A gap analysis provides a look at where the technology needs work in order to meet the needs of LLNL. Finally, recommendations are made to meet these needs.

None

Locating specific 3D objects in overhead images is an important problem in many remote sensing applications. 3D objects may contain either one connected component or multiple disconnected components. Solutions must accommodate images acquired with diverse sensors at various times of the day, in various seasons of the year, or under various weather conditions. Moreover, the physical manifestation of a 3D object with fixed physical dimensions in an overhead image is highly dependent on object physical dimensions, object position/orientation, image spatial resolution, and imaging geometry (e.g., obliqueness). This paper describes a two-stage computer-assisted approach for locating 3D objects in overhead images. In the matching stage, the computer matches models of 3D objects to overhead images. The strongest degree of match over all object orientations is computed at each pixel. Unambiguous local maxima in the degree of match as a function of pixel location are then found. In the cueing stage, the computer sorts image thumbnails in descending order of figure-of-merit and presents them to human analysts for visual inspection and interpretation. The figure-of-merit associated with an image thumbnail is computed from the degrees of match to a 3D object model associated with unambiguous local maxima that lie within the thumbnail. This …

Making solid energetic materials requires the physical mixing of solid fuels and oxidizers or the incorporation of fuel and oxidizing moieties into a single molecule. The former are referred to as composite energetic materials (i.e., thermites, propellants, pyrotechnics) and the latter are deemed monomolecular energetic materials (i.e., explosives). Mass diffusion between the fuel and oxidizer is the rate controlling step for composite reactions while bond breaking and chemical kinetics control monomolecular reactions. Although composites have higher energy densities than monomolecular species, they release that energy over a longer period of time because diffusion controlled reactions are considerably slower than chemistry controlled reactions. Conversely, monomolecular species exhibit greater power due to more rapid kinetics than physically mixed energetics. Reducing the diffusion distance between fuel and oxidizer species within an energetic composite would enhance the reaction rate. Recent advances in nanotechnology have spurred the development of nano-scale fuel and oxidizer particles that can be combined into a composite and effectively reduce diffusion distances to nano-scale dimensions or less. These nanocomposites have the potential to deliver the best of both worlds: high energy density of the physically mixed composite with the high power of the monomolecular species. Toward this end, researchers at Lawrence …

Our first goal was to produce laboratory-scale test beds with a four-probe electrical impedance instrument having 32 current sources. The first test bed, filled with saline, resembles a larger lab-sized test bed at Lawrence Livermore National Lab. We studied arrays of four rods, each with 8 electrodes in this saline-filled tank. We then placed these rods in a test bed of Ottawa sand, and studied the conductivity of a saturated and unsaturated saline/sand bed. This test bed resembles the field test site in the ground at LLNL. The LLNL lab and the field site have only a single current source. Using our existing 32-source instrument, we studied an array of four 8-electrode probes in both the saline and sand tanks. The results were include in our report to you last year, and are summarized below. Figure 1 is a diagram of the four electrode rods, as seen from above, showing seven positions at which a small conductive target was placed, at the level of the middle of the rods. Figure 2 shows the sensitivity of the system to the target presence, as measured by norm distinguishability, at each of these seven positions. The distinguishability was substantially higher when the optimal, …