We derive new self-consistent theoretical UV, optical, and IR diagnostics for the ISM pressure and electron density in the ionized nebulae of star-forming galaxies. Our UV diagnostics utilize the intercombination, forbidden and resonance lines of silicon, carbon, aluminum, neon, and nitrogen. We also calibrate the optical and IR forbidden lines of oxygen, argon, nitrogen and sulfur. We show that line ratios used as ISM pressure diagnostics depend on the gas-phase metallicity with a residual dependence on the ionization parameter of the gas. In addition, the traditional electron density diagnostic [S II] λ6731/[S II] λ6717 is strongly dependent on the gas-phase metallicity. We show how different emission-line ratios are produced in different ionization zones in our theoretical nebulae. The [S II] and [O II] ratios are produced in different zones, and should not be used interchangeably to measure the electron density of the gas unless the electron temperature is known to be constant. We review the temperature and density distributions observed within H II regions and discuss the implications of these distributions on measuring the electron density of the gas. Many H II regions contain radial variations in density. We suggest that the ISM pressure is a more meaningful quantity to measure in H II regions or galaxies. Specific combinations of line ratios can cover the full range of ISM pressures (4 < log(P/k) < 9). As H II regions become resolved at increasingly high redshift through the next generation telescopes, we anticipate that these diagnostics will be important for understanding the conditions around the young, hot stars from the early universe to the present day.

Cluster ensembles combine multiple clusterings of a set of objects into a single consolidated clustering, often referred to as the consensus solution. Consensus clustering can be used to generate more robust and stable clustering results compared to a single clustering approach, perform distributed computing under privacy or sharing constraints, or reuse existing knowledge. This paper describes a variety of algorithms that have been proposed to address the cluster ensemble problem, organizing them in conceptual categories that bring out the common threads and lessons learnt while simultaneously highlighting unique features of individual approaches. © 2011 John Wiley & Sons, Inc. WIREs Data Mining Knowl Discov 2011 1 305–315 DOI: 10.1002/widm.32This article is categorized under: Technologies > Structure Discovery and Clustering

Abstract. The combination of multiple classifiers to generate a single classifier has been shown to be very useful in practice. Similarly, several efforts have shown that cluster ensembles can improve the quality of results as compared to a single clustering solution. These observations suggest that ensembles containing both classifiers and clusterers are potentially useful as well. Specifically, clusterers provide supplementary constraints that can improve the generalization capability of the resulting classifier. This paper introduces a new algorithm named C 3 E that combines ensembles of classifiers and clusterers. Our experimental evaluation of C 3 E shows that it provides good classification accuracies in eleven tasks derived from three real-world applications. In addition, C 3 E produces better results than the recently introduced Bipartite Graphbased Consensus Maximization (BGCM) Algorithm, which combines multiple supervised and unsupervised models and is the algorithm most closely related to C 3 E.

Abstract. The task of matrix completion involves estimating the entries of a matrix, M ∈ R m×n , when a subset, Ω ⊂ {(i, j) : 1 ≤ i ≤ m, 1 ≤ j ≤ n} of the entries are observed. A particular set of low rank models for this task approximate the matrix as a product of two low rank matrices, M = U V T , where U ∈ R m×k and V ∈ R n×k and k min{m, n}. A popular algorithm of choice in practice for recovering M from the partially observed matrix using the low rank assumption is alternating least square (ALS) minimization, which involves optimizing over U and V in an alternating manner to minimize the squared error over observed entries while keeping the other factor fixed. Despite being widely experimented in practice, only recently were theoretical guarantees established bounding the error of the matrix estimated from ALS to that of the original matrix M . In this work we extend the results for a noiseless setting and provide the first guarantees for recovery under noise for alternating minimization. We specifically show that for well conditioned matrices corrupted by random noise of bounded Frobenius norm, if the number of observed entries is O k 7 n log n , then the ALS algorithm recovers the original matrix within an error bound that depends on the norm of the noise matrix. The sample complexity is the same as derived in [7] for the noise-free matrix completion using ALS.

An optimization framework for combining ensembles of classifiers and clusterers with applications to nontransductive semisupervised learning and transfer learning. Data, New York, v.9, n.1, p Unsupervised models can provide supplementary soft constraints to help classify new "target" data because similar instances in the target set are more likely to share the same class label. Such models can also help detect possible differences between training and target distributions, which is useful in applications where concept drift may take place, as in transfer learning settings. This article describes a general optimization framework that takes as input class membership estimates from existing classifiers learned on previously encountered "source" (or training) data, as well as a similarity matrix from a cluster ensemble operating solely on the target (or test) data to be classified, and yields a consensus labeling of the target data. More precisely, the application settings considered are nontransductive semisupervised and transfer learning scenarios where the training data are used only to build an ensemble of classifiers and are subsequently discarded before classifying the target data. The framework admits a wide range of loss functions and classification/clustering methods. It exploits properties of Bregman divergences in conjunction with Legendre duality to yield a principled and scalable approach. A variety of experiments show that the proposed framework can yield results substantially superior to those provided by naïvely applying classifiers learned on the original task to the target data. In addition, we show that the proposed approach, even not being conceptually transductive, can provide better results compared to some popular transductive learning techniques. ACM Transactions on Knowledge Discovery from