ARTMAP: Supervised Real-Time Learning and Classification of Nonstationary Data by a Self-Organizing Neural Network

This article introduces a new neural network architecture, called ARTMAP, that autonomously learns to classify arbitrarily many, arbitrarily ordered vectors into recognition categories based on predictive success. This supervised learning system is built up from a pair of Adaptive Resonance Theory modules (ARTa and ARTb) that are capable of self-organizing stable recognition categories in response to arbitrary sequences of input patterns. During training trials, the ARTa module receives a stream {a^(p)} of input patterns, and ARTb receives a stream {b^(p)} of input patterns, where b^(p) is the correct prediction given a^(p). These ART modules are linked by an associative learning network and an internal controller that ensures autonomous system operation in real time. During test trials, the remaining patterns a^(p) are presented without b^(p), and their predictions at ARTb are compared with b^(p). Tested on a benchmark machine learning database in both on-line and off-line simulations, the ARTMAP system learns orders of magnitude more quickly, efficiently, and accurately than alternative algorithms, and achieves 100% accuracy after training on less than half the input patterns in the database. It achieves these properties by using an internal controller that conjointly maximizes predictive generalization and minimizes predictive error by linking predictive success to category size on a trial-by-trial basis, using only local operations. This computation increases the vigilance parameter ρa of ARTa by the minimal amount needed to correct a predictive error at ARTb· Parameter ρa calibrates the minimum confidence that ARTa must have in a category, or hypothesis, activated by an input a^(p) in order for ARTa to accept that category, rather than search for a better one through an automatically controlled process of hypothesis testing. Parameter ρa is compared with the degree of match between a^(p) and the top-down learned expectation, or prototype, that is read-out subsequent to activation of an ARTa category. Search occurs if the degree of match is less than ρa. ARTMAP is hereby a type of self-organizing expert system that calibrates the selectivity of its hypotheses based upon predictive success. As a result, rare but important events can be quickly and sharply distinguished even if they are similar to frequent events with different consequences. Between input trials ρa relaxes to a baseline vigilance pa When ρa is large, the system runs in a conservative mode, wherein predictions are made only if the system is confident of the outcome. Very few false-alarm errors then occur at any stage of learning, yet the system reaches asymptote with no loss of speed. Because ARTMAP learning is self stabilizing, it can continue learning one or more databases, without degrading its corpus of memories, until its full memory capacity is utilized.

Baseline health-related quality-of-life data as prognostic factors in a phase III multicentre study of women with metastatic breast cancer.

The potential value of baseline health-related quality-of-life (HRQOL) and clinical factors in predicting prognosis was examined using data from an international randomised phase III trial which compared doxorubicin and paclitaxel with doxorubicin and cylophosphamide as first line chemotherapy in 275 women with metastatic breast cancer. The European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 and the related breast module (QLQ-BR23) were used to assess baseline HRQOL data. The Cox proportional-hazards regression model was used for both univariate and multivariate analyses of survival. In the univariate analyses, performance status (P<0.001) and number of sites involved (P=0.001) were the most important clinical prognostic factors. The HRQOL variables at baseline most strongly associated with longer survival were better appetite, physical and role functioning, as well as less fatigue (P<0.001). The final multivariate model retained performance status (P<0.001) and appetite loss (P=0.005) as the variables best predicting survival. Substantial loss of appetite was the only independent HRQOL factor predicting poor survival and was strongly correlated (/r/>0.5) with fatigue, role and physical functioning. In addition to known clinical factors, appetite loss appears to be a significant prognostic factor for survival in women with metastatic breast cancer. However, the mechanism underlying this association remains to be precisely defined in future studies.

Examining Relationships between Culture, Creativity and Business Stage in an Emerging Market: A Categorical Data Analysis of Vietnam’s Data Set

In this article, we offer a new way of exploring relationships between three different dimensions of a business operation, namely the stage of business development, the methods of creativity and the major cultural values. Although separately, each of these has gained enormous attention from the management research community, evidenced by a large volume of research studies, there have been not many studies that attempt to describe the logic that connect these three important aspects of a business; let alone empirical evidences that support any significant relationships among these variables. The paper also provides a data set and an empirical investigation on that data set, using a categorical data analysis, to conclude that examinations of these possible relationships are meaningful and possible for seemingly unquantifiable information. The results also show that the most significant category among all creativity methods employed in Vietnamese enterprises is the “creative disciplines” rule in the “entrepreneurial phase,” while in general creative disciplines have played a critical role in explaining the structure of our data sample, for both stages of development in our consideration.

Optimized approach to decision fusion of heterogeneous data for breast cancer diagnosis.

As more diagnostic testing options become available to physicians, it becomes more difficult to combine various types of medical information together in order to optimize the overall diagnosis. To improve diagnostic performance, here we introduce an approach to optimize a decision-fusion technique to combine heterogeneous information, such as from different modalities, feature categories, or institutions. For classifier comparison we used two performance metrics: The receiving operator characteristic (ROC) area under the curve [area under the ROC curve (AUC)] and the normalized partial area under the curve (pAUC). This study used four classifiers: Linear discriminant analysis (LDA), artificial neural network (ANN), and two variants of our decision-fusion technique, AUC-optimized (DF-A) and pAUC-optimized (DF-P) decision fusion. We applied each of these classifiers with 100-fold cross-validation to two heterogeneous breast cancer data sets: One of mass lesion features and a much more challenging one of microcalcification lesion features. For the calcification data set, DF-A outperformed the other classifiers in terms of AUC (p < 0.02) and achieved AUC=0.85 +/- 0.01. The DF-P surpassed the other classifiers in terms of pAUC (p < 0.01) and reached pAUC=0.38 +/- 0.02. For the mass data set, DF-A outperformed both the ANN and the LDA (p < 0.04) and achieved AUC=0.94 +/- 0.01. Although for this data set there were no statistically significant differences among the classifiers' pAUC values (pAUC=0.57 +/- 0.07 to 0.67 +/- 0.05, p > 0.10), the DF-P did significantly improve specificity versus the LDA at both 98% and 100% sensitivity (p < 0.04). In conclusion, decision fusion directly optimized clinically significant performance measures, such as AUC and pAUC, and sometimes outperformed two well-known machine-learning techniques when applied to two different breast cancer data sets.

Predicting the frequency dispersion of electronic hyperpolarizabilities on the basis of absorption data and thomas-kuhn sum rules

Successfully predicting the frequency dispersion of electronic hyperpolarizabilities is an unresolved challenge in materials science and electronic structure theory. We show that the generalized Thomas-Kuhn sum rules, combined with linear absorption data and measured hyperpolarizability at one or two frequencies, may be used to predict the entire frequency-dependent electronic hyperpolarizability spectrum. This treatment includes two- and three-level contributions that arise from the lowest two or three excited electronic state manifolds, enabling us to describe the unusual observed frequency dispersion of the dynamic hyperpolarizability in high oscillator strength M-PZn chromophores, where (porphinato)zinc(II) (PZn) and metal(II)polypyridyl (M) units are connected via an ethyne unit that aligns the high oscillator strength transition dipoles of these components in a head-to-tail arrangement. We show that some of these structures can possess very similar linear absorption spectra yet manifest dramatically different frequency dependent hyperpolarizabilities, because of three-level contributions that result from excited state-to excited state transition dipoles among charge polarized states. Importantly, this approach provides a quantitative scheme to use linear optical absorption spectra and very limited individual hyperpolarizability measurements to predict the entire frequency-dependent nonlinear optical response. Copyright © 2010 American Chemical Society.

Structuring the Calibration of Qualitative Data as Sets for Qualitative Comparative Analysis (QCA)

Most studies that apply qualitative comparative analysis (QCA) rely on macro-level data, but an increasing number of studies focus on units of analysis at the micro or meso level (i.e., households, firms, protected areas, communities, or local governments). For such studies, qualitative interview data are often the primary source of information. Yet, so far no procedure is available describing how to calibrate qualitative data as fuzzy sets. The authors propose a technique to do so and illustrate it using examples from a study of Guatemalan local governments. By spelling out the details of this important analytic step, the authors aim at contributing to the growing literature on best practice in QCA. © The Author(s) 2012.

Iterative inversion of experimental data in weighted spaces

We present iterative algorithms for solving linear inverse problems with discrete data and compare their performances with the method of singular function expansion, in view of applications in optical imaging and particle sizing.

Particle sizing by inversion of extinction data

info:eu-repo/semantics/published

Vibrational spectra and crystal structure of the di-amino acid peptide cyclo(L-Met-L-Met): Comparison of experimental data and DFT calculations

Experimental Raman and FT-IR spectra of solid-state non-deuterated and N-deuterated samples of cyclo(L-Met-L-Met) are reported and discussed. The Raman and FT-IR results show characteristic amide I vibrations (Raman: 1649 cm-1, infrared: 1675 cm-1) for molecules exhibiting a cis amide conformation. A Raman band, assigned to the cis amide II vibrational mode, is observed at sim1493 cm-1 but no IR band is observed in this region. Cyclo(L-Met-L-Met) crystallises in the triclinic space group P1 with one molecule per unit cell. The overall shape of the diketopiperazine (DKP) ring displays a (slightly distorted) boat conformation. The crystal packing employs two strong hydrogen bonds, which traverse the entire crystal via translational repeats. B3-LYP/cc-pVDZ calculations of the structure of the molecule predict a boat conformation for the DKP ring, in agreement with the experimentally determined X-ray structure. Copyright © 2009 John Wiley & Sons, Ltd.

Use of satellite data for modelling food availability and survival of marine fish larvae

A modelling scheme is described which uses satellite retrieved sea-surface temperature and chlorophyll-a to derive monthly zooplankton biomass estimates in the eastern North Atlantic; this forms part of a bio-physical model of inter-annual variations in the growth and survival of larvae and post-larvae of mackerel (Scomber scombrus). The temperature and chlorophyll data are incorporated first to model copepod (Calanus) egg production rates. Egg production is then converted to available food using distribution data from the Continuous Plankton Recorder (CPR) Survey, observed population biomass per unit daily egg production and the proportion of the larval mackerel diet comprising Calanus. Results are validated in comparison with field observations of zooplankton biomass. The principal benefit of the modelling scheme is the ability to use the combination of broad scale coverage and fine scale temporal and spatial variability of satellite data as driving forces in the model; weaknesses are the simplicity of the egg production model and the broad-scale generalizations assumed in the raising factors to convert egg production to biomass.

History of the Continuous Plankton Recorder Database. Proceedings of 'The Colour of Ocean Data' Symposium, Brussels, 25-27 November 2002.

The Continuous Plankton Recorder Survey has operated in the North Atlantic and North Sea since 1931, providing a unitque multi-decadal dataset of plankton abundance. Over the period since 1931 technology has advanced and the system for storing the CPR data has developed considerably. From 1969 an electronic database was developed to store the results of CPR analysis. Since that time the CPR database has undergone a number of changes due to performance related factors such as processor speed and disk capacity as well as economic factors such as the cost of software. These issues have been overcome and the system for storing and retrieving the data has become more user friendly at every development stage.