Search

We obtained dual-polarization VLBI observations at six frequencies of the compact symmetric object J0029+3457 and the CSO candidate J1324+4048. By comparing the three lower-frequency maps with extrapolations of the high frequency maps we produced maps of the optical depth as a function of frequency. The morphology of the optical-depth maps of J1324+4048 is strikingly smooth, suggestive of a foreground screen of absorbing gas. The spectra at the intensity peaks fit a simple free-free absorption model, with a reduced chi square ~ 2, better than a simple synchrotron self-absorption model, in which the reduced chi square ~ 3.5 - 5.5. We conclude that the case for free-free absorption in J1324+4048 is strong. The optical-depth maps of J0029+3457 exhibit structure, but the morphology does not correlate with that in the intensity maps. The fit of the spectra at the peaks to a simple free-free absorption model yields a reduced chi square ~ 1, but since the turnover is gradual the fit is relatively insensitive to the input parameters. We find that free-free absorption by a thin amount of gas in J0029+3457 is likely, but not definitive. One compact feature in J0029+3457 has an inverted spectrum even at the highest frequencies. We infer this to be the location of the core and estimate an upper limit to the magnetic field of order 3 Gauss at a radius of order 1 pc. In comparison with maps from observations at earlier epochs, no apparent growth in either J1324+4048 or J0029+3457 is apparent, with upper limits of 0.03 and 0.02 mas per yr, corresponding to maximum linear separation speeds of 0.6c and 0.4c., Comment: 31 pages, 8 figures, 5 tables. To be published in the Astrophysical Journal (Accepted: 21-November-2013)

Electronic health records contain rich textual data which possess critical predictive information for machine-learning based diagnostic aids. However many traditional machine learning methods fail to simultaneously integrate both vector space data and text. We present a supervised method using Laplacian eigenmaps to augment existing machine-learning methods with low-dimensional representations of textual predictors which preserve the local similarities. The proposed implementation performs alternating optimization using gradient descent. For the evaluation we applied our method to over 2,000 patient records from a large single-center pediatric cardiology practice to predict if patients were diagnosed with cardiac disease. Our method was compared with latent semantic indexing, latent Dirichlet allocation, and local Fisher discriminant analysis. The results were assessed using AUC, MCC, specificity, and sensitivity. Results indicate supervised Laplacian eigenmaps was the highest performing method in our study, achieving 0.782 and 0.374 for AUC and MCC respectively. SLE showed an increase in 8.16% in AUC and 20.6% in MCC over the baseline which excluded textual data and a 2.69% and 5.35% increase in AUC and MCC respectively over unsupervised Laplacian eigenmaps. This method allows many existing machine learning predictors to effectively and efficiently utilize the potential of textual predictors.

Computational Electromagnetic Software is used by NASA to analyze the compatibility of systems too large or too complex for testing. Recent advances in software packages and computer capabilities have made it possible to determine the effects of a transmitter inside a launch vehicle fairing, better analyze the environment threats, and perform on-orbit replacements with assured electromagnetic compatibility.