Analysis of multivariate stochastic signals sampled by on-line particle analyzers: Application to the quantitative assessment of occupational exposure to NOAA in multisource industrial scenarios (MSIS)

In multisource industrial scenarios (MSIS) coexist NOAA generating activities with other productive sources of airborne particles, such as parallel processes of manufacturing or electrical and diesel machinery. A distinctive characteristic of MSIS is the spatially complex distribution of aerosol sources, as well as their potential differences in dynamics, due to the feasibility of multi-task configuration at a given time. Thus, the background signal is expected to challenge the aerosol analyzers at a probably wide range of concentrations and size distributions, depending of the multisource configuration at a given time. Monitoring and prediction by using statistical analysis of time series captured by on-line particle analyzers in industrial scenarios, have been proven to be feasible in predicting PNC evolution provided a given quality of net signals (difference between signal at source and background). However the analysis and modelling of non-consistent time series, influenced by low levels of SNR (Signal-Noise Ratio) could build a misleading basis for decision making. In this context, this work explores the use of stochastic models based on ARIMA methodology to monitor and predict exposure values (PNC). The study was carried out in a MSIS where an case study focused on the manufacture of perforated tablets of nano-TiO2 by cold pressing was performed

Marine Fisheries Stock Assessment Improvement Plan: report of the National Marine Fisheries Service National Task Force for Improving Fish Stock Assessments

This report argues for greatly increased resources in terms of data collection facilities and staff to collect, process, and analyze the data, and to communicate the results, in order for NMFS to fulfill its mandate to conserve and manage marine resources. In fact, the authors of this report had great difficulty defining the "ideal" situation to which fisheries stock assessments and management should aspire. One of the primary objectives of fisheries management is to develop sustainable harvest policies that minimize the risks of overfishing both target species and associated species. This can be achieved in a wide spectrum of ways, ranging between the following two extremes. The first is to implement only simple management measures with correspondingly simple assessment demands, which will usually mean setting fishing mortality targets at relatively low levels in order to reduce the risk of unknowingly overfishing or driving ecosystems towards undesirable system states. The second is to expand existing data collection and analysis programs to provide an adequate knowledge base that can support higher fishing mortality targets while still ensuring low risk to target and associated species and ecosystems. However, defining "adequate" is difficult, especially when scientists have not even identified all marine species, and information on catches, abundances, and life histories of many target species, and most associated species, is sparse. Increasing calls from the public, stakeholders, and the scientific community to implement ecosystem-based stock assessment and management make it even more difficult to define "adequate," especially when "ecosystem-based management" is itself not well-defined. In attempting to describe the data collection and assessment needs for the latter, the authors took a pragmatic approach, rather than trying to estimate the resources required to develop a knowledge base about the fine-scale detailed distributions, abundances, and associations of all marine species. Thus, the specified resource requirements will not meet the expectations of some stakeholders. In addition, the Stock Assessment Improvement Plan is designed to be complementary to other related plans, and therefore does not duplicate the resource requirements detailed in those plans, except as otherwise noted.

Incremental Linear Discriminant Analysis Using Sufficient Spanning Sets and Its Applications

This paper presents an incremental learning solution for Linear Discriminant Analysis (LDA) and its applications to object recognition problems. We apply the sufficient spanning set approximation in three steps i.e. update for the total scatter matrix, between-class scatter matrix and the projected data matrix, which leads an online solution which closely agrees with the batch solution in accuracy while significantly reducing the computational complexity. The algorithm yields an efficient solution to incremental LDA even when the number of classes as well as the set size is large. The incremental LDA method has been also shown useful for semi-supervised online learning. Label propagation is done by integrating the incremental LDA into an EM framework. The method has been demonstrated in the task of merging large datasets which were collected during MPEG standardization for face image retrieval, face authentication using the BANCA dataset, and object categorisation using the Caltech101 dataset. © 2010 Springer Science+Business Media, LLC.

Vibration and stress analysis in the presence of structural uncertainty

At medium to high frequencies the dynamic response of a built-up engineering system, such as an automobile, can be sensitive to small random manufacturing imperfections. Ideally the statistics of the system response in the presence of these uncertainties should be computed at the design stage, but in practice this is an extremely difficult task. In this paper a brief review of the methods available for the analysis of systems with uncertainty is presented, and attention is then focused on two particular "non- parametric" methods: statistical energy analysis (SEA), and the hybrid method. The main governing equations are presented, and a number of example applications are considered, ranging from academic benchmark studies to industrial design studies. © 2009 IOP Publishing Ltd.

Factor analysis based VTS and JUD noise estimation and compensation

Model based compensation schemes are a powerful approach for noise robust speech recognition. Recently there have been a number of investigations into adaptive training, and estimating the noise models used for model adaptation. This paper examines the use of EM-based schemes for both canonical models and noise estimation, including discriminative adaptive training. One issue that arises when estimating the noise model is a mismatch between the noise estimation approximation and final model compensation scheme. This paper proposes FA-style compensation where this mismatch is eliminated, though at the expense of a sensitivity to the initial noise estimates. EM-based discriminative adaptive training is evaluated on in-car and Aurora4 tasks. FA-style compensation is then evaluated in an incremental mode on the in-car task. © 2011 IEEE.

Identification of native American founder mtDNAs through the analysis of complete mtDNA sequences: Some caveats

In this study, a detailed analysis of both previously published and new data was performed to determine whether complete, or almost complete, mtDNA sequences can resolve the long-debated issue of which Asian mtDNAs were founder sequences for the Native American mtDNA pool. Unfortunately, we now know that coding region data and their analysis are not without problems. To obtain and report reasonably correct sequences does not seem to be a trivial task, and to discriminate between Asian-and Native American mtDNA ancestries may be more complex than previously believed. It is essential to take into account the effects of mutational hot spots in both the control and coding regions, so that the number of apparent Native American mtDNA founder sequences is not erroneously inflated. As we report here, a careful analysis of all available data indicates that there is very little evidence that more than five founder mtDNA sequences entered Beringia before the Last Glacial Maximum and left their traces in the current Native American mtDNA pool.

Evaluating the impacts of iteration on PD processes by transforming task network models into System Dynamics models

Iteration is unavoidable in the design process and should be incorporated when planning and managing projects in order to minimize surprises and reduce schedule distortions. However, planning and managing iteration is challenging because the relationships between its causes and effects are complex. Most approaches which use mathematical models to analyze the impact of iteration on the design process focus on a relatively small number of its causes and effects. Therefore, insights derived from these analytical models may not be robust under a broader consideration of potential influencing factors. In this article, we synthesize an explanatory framework which describes the network of causes and effects of iteration identified from the literature, and introduce an analytic approach which combines a task network modeling approach with System Dynamics simulation. Our approach models the network of causes and effects of iteration alongside the process architecture which is required to analyze the impact of iteration on design process performance. We show how this allows managers to assess the impact of changes to process architecture and to management levers which influence iterative behavior, accounting for the fact that these changes can occur simultaneously and can accumulate in non-linear ways. We also discuss how the insights resulting from this analysis can be visualized for easier consumption by project participants not familiar with simulation methods. Copyright © 2010 by ASME.

Grand Challenges for Data Sensing and Analysis

A diverse group of experts proposed the 9 grand challenges outlined in this booklet. This expert task force was assembled by the ASCE TCCIT Data Sensing and Analysis (DSA) Committee and endorsed by the TRB AFH10(1) Construction IT joint subcommittee at the request of their membership. The task force did not rank the challenges selected, nor did it endorse particular approaches to meeting them. Rather than attempt to include every important goal for data sensing and analysis, the panel chose opportunities that were both achievable and sustainable to help people and the planet thrive. The panel’s conclusions were reviewed by several subject-matter experts. The DSA is offering an opportunity to comment on the challenges by contacting the task force chair via email at becerik@usc.edu.

Factor analysis based VTS discriminative adaptive training

Vector Taylor Series (VTS) model based compensation is a powerful approach for noise robust speech recognition. An important extension to this approach is VTS adaptive training (VAT), which allows canonical models to be estimated on diverse noise-degraded training data. These canonical model can be estimated using EM-based approaches, allowing simple extensions to discriminative VAT (DVAT). However to ensure a diagonal corrupted speech covariance matrix the Jacobian (loading matrix) relating the noise and clean speech is diagonalised. In this work an approach for yielding optimal diagonal loading matrices based on minimising the expected KL-divergence between the diagonal loading matrix and "correct" distributions is proposed. The performance of DVAT using the standard and optimal diagonalisation was evaluated on both in-car collected data and the Aurora4 task. © 2012 IEEE.

Generalized power method for sparse principal component analysis

In this paper we develop a new approach to sparse principal component analysis (sparse PCA). We propose two single-unit and two block optimization formulations of the sparse PCA problem, aimed at extracting a single sparse dominant principal component of a data matrix, or more components at once, respectively. While the initial formulations involve nonconvex functions, and are therefore computationally intractable, we rewrite them into the form of an optimization program involving maximization of a convex function on a compact set. The dimension of the search space is decreased enormously if the data matrix has many more columns (variables) than rows. We then propose and analyze a simple gradient method suited for the task. It appears that our algorithm has best convergence properties in the case when either the objective function or the feasible set are strongly convex, which is the case with our single-unit formulations and can be enforced in the block case. Finally, we demonstrate numerically on a set of random and gene expression test problems that our approach outperforms existing algorithms both in quality of the obtained solution and in computational speed. © 2010 Michel Journée, Yurii Nesterov, Peter Richtárik and Rodolphe Sepulchre.

On Combining Morphological Component Analysis and Concentric Morphology Model for Mammographic Mass Detection

Mammographic mass detection is an important task for the early diagnosis of breast cancer. However, it is difficult to distinguish masses from normal regions because of their abundant morphological characteristics and ambiguous margins. To improve the mass detection performance, it is essential to effectively preprocess mammogram to preserve both the intensity distribution and morphological characteristics of regions. In this paper, morphological component analysis is first introduced to decompose a mammogram into a piecewise-smooth component and a texture component. The former is utilized in our detection scheme as it effectively suppresses both structural noises and effects of blood vessels. Then, we propose two novel concentric layer criteria to detect different types of suspicious regions in a mammogram. The combination is evaluated based on the Digital Database for Screening Mammography, where 100 malignant cases and 50 benign cases are utilized. The sensitivity of the proposed scheme is 99% in malignant, 88% in benign, and 95.3% in all types of cases. The results show that the proposed detection scheme achieves satisfactory detection performance and preferable compromises between sensitivity and false positive rates.

New method for analysis of Chinese herbal complex prescription and its application

Traditional Chinese medicine (TCM) is a great treasure of China, the analysis of which is an arduous task. The viewpoint that all chemical constituents of Chinese herbal complex prescription should be analyzed as a black box is elucidated for the first time. Intelligent multi-mode multi-column chromatographic system (IMMCC) with its hybrids is the basic method and HPLC Unified Method is the breakthrough for the black box analysis. Dang-Gui-Bu-Xue-Tang was selected as a typical TCM and a systematic separation method from non-aqueous mobile phase to pure water mobile phase was put forward in order to convert unknown sample to known sample. The a, c values and UV spectra of 66 components of Astragalus, 78 components of Angelica and 71 components of Dang-Gui-Bu-Xue-Tang were obtained. Intelligent optimization and peak identification method and software for complex samples were developed and the optimum multi-step multi-binary gradient curve of mobile phase for Astragalus was ascertained. The maximum error and minimum error of predicted retention time for all components of Astragalus are 8.62% and 0.05% respectively. All components of Astragalus were compared with those of Angelica and it is found that many components of Astragalus are the same as those of Angelica, while the contents of these components are different. Many components of Dang-Gui-Bu-Xue-Tang are also the same as those of Astragalus and Angelica with different contents.

Theories of Comparative Analysis

Comparative analysis is the problem of predicting how a system will react to perturbations in its parameters, and why. For example, comparative analysis could be asked to explain why the period of an oscillating spring/block system would increase if the mass of the block were larger. This thesis formalizes the task of comparative analysis and presents two solution techniques: differential qualitative (DQ) analysis and exaggeration. Both techniques solve many comparative analysis problems, providing explanations suitable for use by design systems, automated diagnosis, intelligent tutoring systems, and explanation based generalization. This thesis explains the theoretical basis for each technique, describes how they are implemented, and discusses the difference between the two. DQ analysis is sound; it never generates an incorrect answer to a comparative analysis question. Although exaggeration does occasionally produce misleading answers, it solves a larger class of problems than DQ analysis and frequently results in simpler explanations.

Automatic diagnosis of task faults in flexible manufacturing systems

Hardy, N. W., Barnes, D. P., Lee, L. H. (1989). Automatic diagnosis of task faults in flexible manufacturing systems. Robotica, 7 (1):25-35

Qualitative Circuit Models in Failure Analysis Reasoning

Lee M.H., Qualitative Circuit Models in Failure Analysis Reasoning, AI Journal. vol 111, pp239-276.1999.