Automatic Determination of Stellar Atmospheric Parameters Using Neural Networks and Instance-Based Machine Learning

In this article we show how machine learning methods can beeffectively applied to the problem of automatically predictingstellar atmospheric parameters from spectral information, a veryimportant problem in stellar astronomy. We apply feedforwardneural networks, Kohonen's self-organizing maps andlocally-weighted regression to predict the stellar atmosphericparameters effective temperature, surface gravity and metallicityfrom spectral indices. Our experimental results show that thethree methods are capable of predicting the parameters with verygood accuracy. Locally weighted regression gives slightly betterresults than the other methods using the original dataset asinput, while self-organizing maps outperform the other methods when significant amounts of noise are added. We also implemented a heterogeneous ensemble of predictors, combining the results given by the three algorithms. This ensemble yields better results than any of the three algorithms alone, using both the original and the noisy data.     

A chum scale pattern analyzer and stock identification based onscale patterns by discriminant function and power spectrum analysis

An optoelectronic instrument has been developed to analyze the scales and scale patterns of chum salmon. This device, with specially developed software, has been used to study populations of salmon. Fish from two different rivers approximately 200 km apart from each other were grouped with an accuracy of 82% or more. Machine-measured ages coincided with those derived from visual estimation in approximately 80% of the fish sampled. These results suggest the proposed system could be of practical use for measuring and analyzing the scale patterns that allow one to accurately measure both natural and farmed populations of salmon     

Examples of clustering around high redshift radio-galaxies

Published in Astrofizika, Vol. 38, No. 4, pp. 677–680, October–December, 1995.     

Observational Selection of Comet Discoveries

The distribution of perihelion distances of nearly periodic comets was studied using different assumptions about comet discovery probabilities. It was shown that the concentration of perihelia at small heliocentric distances cannot be explained solely by the influence of visibility conditions.     

Accretion disk spectra of super-luminal jet sources and ultra-luminous compact X-ray sources in nearby spiral galaxies

The Ultra-luminous Compact X-ray Sources (ULXs)in nearby spiral galaxies and the Galactic super-luminaljet sources sharethe common spectral characteristic that they haveextremely high disk temperatures which cannot be explainedin the framework of the standard accretion disk modelin the Schwarzschild metric. We have calculated an extreme Kerr disk model to examine if the Kerr disk model can instead explain the observed `too hot' accretion disk spectra.We found that the Kerr disk spectrum becomes significantly hardercompared to the Schwarzschild disk only when the disk is highlyinclined.For super-luminal jet sources, which are known to beinclined systems, the Kerr disk model may thuswork if we choose proper values for the black hole angular momentum. For the ULXs, however, the Kerr disk interpretation will be problematic,as is is highly unlikely that their accretion disks are preferentiallyinclined.     

The Geminid Meteor Stream Activity Profile

The existence of two maxima of the activity of the Geminid meteor stream and the general shape of the stream activity (rate curve) are discussed. The data of visual and radar observations are compared to the results of mathematical simulation. The distribution of the orbits of meteoroids, which are observed on the Earth, is determined from the mathematical model. This distribution cannot as yet be confirmed or disproved because of the absence of appropriate experimental data.     

Starburst activity in a ROSAT narrow emission-line galaxy

We present multiwaveband photometric and optical spectropolarimetric observations of the R =15.9 narrow emission-line galaxy R117_A which lies on the edge of the error circle of the ROSAT X-ray source R117. The overall spectral energy distribution of the galaxy is well modelled by a combination of a normal spiral galaxy and a moderate-strength burst of star formation. The far-infrared and radio emission is extended along the major axis of the galaxy, indicating an extended starburst.
On positional grounds, the galaxy is a good candidate for the identification of R117, and the observed X-ray flux is very close to what would be expected from a starburst of the observed far-infrared and radio fluxes. Although an obscured high-redshift QSO cannot be entirely ruled out as contributing some fraction of the X-ray flux, we find no candidates to K =20.8 within the X-ray error box, and so conclude that R117_A is responsible for a large fraction, if not all, of the X-ray emission from R117.
Searches for indicators of an obscured AGN in R117_A have so far proven negative; deep spectropolarimetric observations show no signs of broad lines to a limit of 1 per cent and, for the observed far-infrared and radio emission, we would expect 10 times greater X-ray flux if the overall emission were powered by an AGN. We therefore conclude that the X-ray emission from R117 is dominated by starburst emission from the galaxy R117_A.     

Abundances of [WC] central stars and their planetary nebulaee

We review elemental abundances derived for planetary nebula (PN) WCcentral stars and for their nebulae. Uncertainties in the abundances of[WC] stars are still too large to enable an abundance sequenceto be constructed. In particular it is not clear why the hotter [WCE]stars have C and O abundances which are systematically lower than those oftheir supposed precursors, the [WCL] stars. This abundance differencecould be real or it may be due to unaccounted-for systematic effects inthe analyses. Hydrogen might not be present in [WC] star winds asoriginallysuggested, since broad pedestals observed at the base of nebular lines canplausibly be attributed to high velocity nebular components. It isrecommended that stellar abundance analyses should be carried out withnon-LTE model codes, although recombination line analyses can provideuseful insights. In particular, C II dielectronic recombinationlines provide a unique means to determine electron temperatures in cool[WC] star winds. We then compare the abundances found for PNe which have [WC] central starswith those that do not. Numerous abundance analyses of PNe have beenpublished, but comparisons based on non-uniform samples and methods arelikely to lack reliability. Nebular C/H ratios, which might be expected todistinguish between PNe around H-poor and H-rich stars, are rather similarfor the two groups, with only a small tendency towards larger values fornebulae around H-deficient stars. Nebular abundances should be obtainedwith photoionization models using the best-fitting non-LTE modelatmosphere for the central star as the input. Heavy-metal line blanketingstill needs to be taken into consideration when modeling the central star,as its omission can significantly affect the ionizing fluxes as well asthe abundance determinations. We discuss the discrepancies between nebularabundances derived from collisionally excited lines and thosederived from optical recombination lines, a phenomenon that may havelinks with the presence of H-deficient central stars.     

Celestial-mechanical peculiarities of Uranus’s satellite system

We consider the structural peculiarities of Uranus’s satellite system associated with its separation into two groups: inner equatorial satellites moving in nearly circular orbits and distant irregular satellites with retrograde motion in highly elliptical orbits. The intermediate region is free from satellites in a wide range of semimajor axes. By analyzing the evolution of satellite orbits under the combined effect of solar attraction and Uranus’s oblateness, we offer a celestial-mechanical explanation for the absence of equatorial satellites in this region. M.L. Lidov’s studies during 1961–1963 have served as a basis for our analysis.     

Noble gases in interplanetary dust particles,II: Excess helium‐3 in cluster particles and modeling constraints on interplanetary dust particle exposures to cosmic‐ray irradiation

Abstract— Measurements of He isotopes in cluster interplanetary dust particles (IDPs) from stratospheric dust collector L2009 reveal anomalous ³He/⁴He ratios comparable to those seen earlier, up to ～40x the solar wind ratio, in particles from the companion collector L2011. These overabundances of ³He in the L2009 samples are masked by much higher ⁴He contents compared to the L2011 particles, and are visible only in minor gas fractions evolved by stepwise heating at high temperatures. Cosmic‐ray induced spallogenic reactions are efficient producers of ³He. The majority of this paper is devoted to a detailed assessment of the possible role of spallation in generating the ³He excesses in these and other cluster IDPs. A model of collisional erosion and fragmentation during inward transit through the interplanetary dust environment is used to estimate space lifetimes of particles from asteroidal and Edgeworth–Kuiper Belt sources. Results of the modeling indicate that Poynting–Robertson orbital evolution timescales of IDPs small enough to elude destruction on their way to Earth from either location are far shorter than the cosmic‐ray exposure ages required to account for observed ³He overabundances. Grains large enough to have sufficiently long space residence times are fragmented close to their sources. An alternative to long in‐space exposure could be prolonged irradiation of particles buried in parent body regoliths prior to their ejection as IDPs. A qualitative calculation suggests, however, that collisional erosion of asteroidal upper‐regolith materials is likely to occur on timescales shorter than the > 1 Ga burial times needed for accumulation of spallogenic ³He to the levels seen in several cluster particles. In contrast, regoliths on Edgeworth–Kuiper Belt objects may be stable enough to account for the ³He excesses, and delivery of heavily pre‐irradiated IDPs to the inner solar system by short‐period Edgeworth–Kuiper Belt comets remains a possibility. A potential problem is that the expected associated abundances of spallation‐produced ²¹Ne appear to be absent, although here the present IDP data base is too sparse and for the most part too imprecise to rule out a spallogenic origin. Relatively short periods of pre‐ejection residence in asteroidal regoliths may be responsible for the curiously broad exposure age distributions reported for micrometeorites extracted from Greenland and sea‐floor sediments.