Class Of Stars Research Articles

Random forest classification of stars in the Galactic Centre

Near-infrared high-angular resolution imaging observations of the Milky Way's nuclear star cluster have revealed all luminous members of the existing stellar population within the central parsec. Generally, these stars are either evolved late-type giants or massive young, early-type stars. We revisit the problem of stellar classification based on intermediate-band photometry in the K-band, with the primary aim of identifying faint early-type candidate stars in the extended vicinity of the central massive black hole. A random forest classifier, trained on a subsample of spectroscopically identified stars, performs similarly well as competitive methods (F1=0.85), without involving any model of stellar spectral energy distributions. Advantages of using such a machine-trained classifier are a minimum of required calibration effort, a predictive accuracy expected to improve as more training data becomes available, and the ease of application to future, larger data sets. By applying this classifier to archive data, we are also able to reproduce the results of previous studies of the spatial distribution and the K-band luminosity function of both the early- and late-type stars.

Classification and Its Consequences for Online Harassment

Online harassment is a pervasive and pernicious problem. Techniques like natural language processing and machine learning are promising approaches for identifying abusive language, but they fail to address structural power imbalances perpetuated by automated labeling and classification. Similarly, platform policies and reporting tools are designed for a seemingly homogenous user base and do not account for individual experiences and systems of social oppression. This paper describes the design and evaluation of HeartMob, a platform built by and for people who are disproportionately affected by the most severe forms of online harassment. We conducted interviews with 18 HeartMob users, both targets and supporters, about their harassment experiences and their use of the site. We examine systems of classification enacted by technical systems, platform policies, and users to demonstrate how 1) labeling serves to validate (or invalidate) harassment experiences; 2) labeling motivates bystanders to provide support; and 3) labeling content as harassment is critical for surfacing community norms around appropriate user behavior. We discuss these results through the lens of Bowker and Star's classification theories and describe implications for labeling and classifying online abuse. Finally, informed by intersectional feminist theory, we argue that fully addressing online harassment requires the ongoing integration of vulnerable users' needs into the design and moderation of online platforms.

The Extremely Luminous Quasar Survey in the SDSS Footprint. I. Infrared-based Candidate Selection

Abstract Studies of the most luminous quasars at high redshift directly probe the evolution of the most massive black holes in the early universe and their connection to massive galaxy formation. However, extremely luminous quasars at high redshift are very rare objects. Only wide-area surveys have a chance to constrain their population. The Sloan Digital Sky Survey (SDSS) has so far provided the most widely adopted measurements of the quasar luminosity function at . However, a careful re-examination of the SDSS quasar sample revealed that the SDSS quasar selection is in fact missing a significant fraction of quasars at the brightest end. We identified the purely optical-color selection of SDSS, where quasars at these redshifts are strongly contaminated by late-type dwarfs, and the spectroscopic incompleteness of the SDSS footprint as the main reasons. Therefore, we designed the Extremely Luminous Quasar Survey (ELQS), based on a novel near-infrared JKW2 color cut using Wide-field Infrared Survey Explorer mission (WISE) AllWISE and 2MASS all-sky photometry, to yield high completeness for very bright ( ) quasars in the redshift range of . It effectively uses random forest machine-learning algorithms on SDSS and WISE photometry for quasar–star classification and photometric redshift estimation. The ELQS will spectroscopically follow-up ∼230 new quasar candidates in an area of ∼12,000 deg2 in the SDSS footprint to obtain a well-defined and complete quasar sample for an accurate measurement of the bright-end quasar luminosity function (QLF) at . In this paper, we present the quasar selection algorithm and the quasar candidate catalog.

Unsupervised classification of variable stars

During the last ten years, a considerable amount of effort has been made to develop algorithms for automatic classification of variable stars. That has been primarily achieved by applying machine learning methods to photometric datasets where objects are represented as light curves. Classifiers require training sets to learn the underlying patterns that allow the separation among classes. Unfortunately, building training sets is an expensive process that demands a lot of human efforts. Every time data comes from new surveys; the only available training instances are the ones that have a cross-match with previously labelled objects, consequently generating insufficient training sets compared with the large amounts of unlabelled sources. In this work, we present an algorithm that performs unsupervised classification of variable stars, relying only on the similarity among light curves. We tackle the unsupervised classification problem by proposing an untraditional approach. Instead of trying to match classes of stars with clusters found by a clustering algorithm, we propose a query based method where astronomers can find groups of variable stars ranked by similarity. We also develop a fast similarity function specific for light curves, based on a novel data structure that allows scaling the search over the entire dataset of unlabelled objects. Experiments show that our unsupervised model achieves high accuracy in the classification of different types of variable stars and that the proposed algorithm scales up to massive amounts of light curves.

NGC 6334 and NGC 6357

Aims. The formation of high-mass stars is still debated. For this reason, several projects such as Herschel-HOBYS are focussed on the study of the earliest phases of massive star formation. As a result, massive star-forming complexes such as NGC 6334 and NGC 6357 have been observed in the far-infrared to study their massive dense cores where massive stars are expected to form. However, to better characterise the environments of these cores we need to understand the previous massive star formation history. To better characterise the environment of these massive dense cores we study the previous high-mass star formation and how these stars act on their environments. Methods. This study is based on the spectral classification of the OB stars identified towards NGC 6334 and NGC 6357 with spectra taken with the AAOmega spectrograph on the Anglo-Australian Telescope (AAT). From the subsequent spectral classification of 109 stars across these regions we were able to evaluate the following: distance, age, mass, global star-forming efficiency (SFE), and star formation rate (SFR) of the regions. The physical conditions of the ionised gas for both complexes was also derived. Results. We confirm that NGC 6334 and NGC 6357 belong to the Saggitarius-Carina arm which, in this direction, extends from 1 kpc to 2.2 kpc. From the location of the stars in Hertzprung-Russell diagram we show that stars older than ∼10 Myr are broadly spread across these complexes, while younger stars are mainly located in the H ii regions and stellar clusters. Our data also suggests that some of the young stars can be considered runaway stars. We evaluate a SFE of 0.019 +0.008 −0.007 and 0.021 +0.004 −0.003 and a SFR of 1.1 × 10 3 ± 300 M Myr −1 and 1.7 × 10 3 ± 400 M Myr −1 for NGC 6334 and NGC 6357, respectively. We note that 29 OB stars have X-ray counterparts, most of them belonging to NGC 6357. This suggests that molecular clouds in NGC 6357 are more impacted by X-ray flux and stellar winds than in NGC 6334. Finally, from the analysis of nebular lines (Hα, [NII], and [SII]) from spectra from several regions of ionised gas, we confirm that the filaments in NGC 6357 are shock heated.

Orbital Dynamics of Candidate Transitional Millisecond Pulsar 3FGL J1544.6-1125: An Unusually Face-on System

Abstract We present the orbital solution for the donor star of the candidate transitional millisecond pulsar (tMSP) 3FGL J1544.6-1125, currently observed as an accreting low-mass X-ray binary. The orbital period is days, entirely consistent with the spectral classification of the donor star as a mid-to-late K dwarf. The semi-amplitude of the radial velocity curve is exceptionally low at km s−1, implying a remarkably face-on inclination in the range 5°–8°, depending on the neutron star and donor masses. After determining the veiling of the secondary, we derive a distance to the binary of 3.8 ± 0.7 kpc, yielding a 0.3–10 keV X-ray luminosity of , similar to confirmed tMSPs. As face-on binaries rarely occur by chance, we discuss the possibility that Fermi-selected samples of transitional millisecond pulsars in the sub-luminous disk state are affected by beaming. By phasing emission line strength on the spectroscopic ephemeris, we find coherent variations, and argue that these variations are most consistent with emission from an asymmetric shock originating near the inner disk.

Photometric Strömgren-Hβ Quantification for O and B Stars of Luminosity Class V

In any photometric system, the relationship between the spectral type and the intrinsic colors is fundamental for the photometric stellar classification and for a reliable derivation of stellar parameters. Such relationships are being constantly verified and improved based on larger and better samples. The capability of the Strömgren-Hβ photometry to provide reddening-free quantities allows the classification of reddened stars over the entire spectral sequence. However, a detailed consistent tabulation of the uvbyβ quantities for the O-B9 part of the main sequence is currently warranted. We combine spectral types and photometry from the literature to provide a new empirical tabulation of the intrinsic color b − y, color differences c1, m1, [c1] and [m1] and the Hβ index for main-sequence (luminosity class V) stars of spectral types O4.5 through B9.5 in terms of spectral sub-types. Our survey is based on more than 1400 stars. We use the established c1 versus b − y intrinsic line and the accepted reddening slopes to obtain dereddened quantities and then calculate their median values as a function of the spectral sub-type. The classification ability of the system in this spectral range is reliable and the photometric quantities are not affected by the stellar rotation.

Automatic Survey-invariant Classification of Variable Stars

Abstract Machine learning techniques have been successfully used to classify variable stars on widely studied astronomical surveys. These data sets have been available to astronomers long enough, thus allowing them to perform deep analysis over several variable sources and generating useful catalogs with identified variable stars. The products of these studies are labeled data that enable supervised learning models to be trained successfully. However, when these models are blindly applied to data from new sky surveys, their performance drops significantly. Furthermore, unlabeled data become available at a much higher rate than their labeled counterpart, since labeling is a manual and time-consuming effort. Domain adaptation techniques aim to learn from a domain where labeled data are available, the source domain, and through some adaptation perform well on a different domain, the target domain. We propose a full probabilistic model that represents the joint distribution of features from two surveys, as well as a probabilistic transformation of the features from one survey to the other. This allows us to transfer labeled data to a study where they are not available and to effectively run a variable star classification model in a new survey. Our model represents the features of each domain as a Gaussian mixture and models the transformation as a translation, rotation, and scaling of each separate component. We perform tests using three different variability catalogs, EROS, MACHO, and HiTS, presenting differences among them, such as the number of observations per star, cadence, observational time, and optical bands observed, among others.

A Comparative Observational Study of YSO Classification in Four Small Star-forming H ii Regions

Abstract We have developed a new young stellar object (YSO) identification and classification technique using mid-infrared Wide-field Infrared Survey Explorer (WISE) data. We compare this new technique with previous WISE YSO detection and classification methods that used either infrared colors or spectral energy distribution slopes. In this study, we also use the new technique to detect and examine the YSO population associated with four small H ii regions: KR 7, KR 81, KR 120, and KR 140. The relatively simple structure of these regions allows us to effectively use both spatial and temporal constraints to identify YSOs that are potential products of triggered star formation. We are also able to identify regions of active star formation around these H ii regions that are clearly not influenced by the H ii region expansion, and thus demonstrate that star formation is on-going on megayear timescales in some of these molecular clouds.

Stars with relativistic speeds in the Hills scenario

The dynamical capture of a binary system consisting of a supermassive black hole (SMBH) and an ordinary star in the gravitational field of a central (more massive) SMBH is considered in the three-body problem in the framework of a modified Hills scenario. The results of numerical simulations predict the existence of objects whose spatial speeds are comparable to the speed of light. The conditions for and constraints imposed on the ejection speeds realized in a classical scenario and the modified Hills scenario are analyzed. The star is modeled using an N-body approach, making it possible to treat it as a structured object, enabling estimation of the probability that the object survives when it is ejected with relativistic speed as a function of the mass of the star, the masses of both SMBHs, and the pericenter distance. It is possible that the modern kinematic classification for stars with anomalously high spatial velocities will be augmented with a new class—stars with relativistic speeds.

Bow shocks in a newly discovered maser source in IRAS 20231+3440

From measuring the annual parallax of water masers over 1.5 yr with VLBI Exploration of Radio Astrometry, we present the trigonometric parallax and corresponding distance of another newly identified water maser source in the region of IRAS 20231+3440 as π = 0.611 ± 0.022 mas and D = 1.64 ± 0.06 kpc, respectively. We measured the absolute proper motions of all the newly detected maser spots (30 spots) and presented two pictures describing the possible spatial distribution of the water maser as the morphology marks out an arc of masers whose average proper motion velocity in the jet direction was 14.26 km s−1. As revealed by the ALLWISE composite image and by applying the colour–colour method of young stellar objects (YSO) identification and classification on photometric archived data, we identified the driving source of the north maser group to be a class I, young stellar object. To further probe the nature of the progenitor, we used the momentum rate maximum value (1.2 × 10−4 M⊙ yr−1 km s−1) of the outflow to satisfy that the progenitor under investigation is a low-mass young stellar object concurrently forming alongside an intermediate-mass YSO ∼60 000 au (∼37 arcsec) away from it.

Pulsating low-mass white dwarfs in the frame of new evolutionary sequences

We present a theoretical assessment of the expected temporal rates of change of periods ($\dot{\Pi}$) for low-mass ($M_{\star}/M_{\sun} \lesssim 0.45$) and extremely low-mass (ELM, $M_{\star}/M_{\sun} \lesssim 0.18-0.20$) white-dwarf stars, based on fully evolutionary low-mass He-core white dwarf and pre-white dwarf models. Our analysis is based on a large set of adiabatic periods of radial and nonradial pulsation modes computed on a suite of low-mass He-core white dwarf and pre-white dwarf models with masses ranging from $0.1554$ to $0.4352 M_{\sun}$. We compute the secular rates of period change of radial ($\ell= 0$) and nonradial ($\ell= 1, 2$) $g$ and $p$ modes for stellar models representative of ELMV and pre-ELMV stars, as well as for stellar objects that are evolving just before the occurrence of CNO flashes at the early cooling branches. We found that the theoretically expected magnitude of $\dot{\Pi}$ of $g$ modes for pre-ELMVs are by far larger than for ELMVs. In turn, $\dot{\Pi}$ of $g$ modes for models evolving before the occurrence of CNO flashes are larger than the maximum values of the rates of period change predicted for pre-ELMV stars. Regarding $p$ and radial modes, we found that the larger absolute values of $\dot{\Pi}$ correspond to pre-ELMV models. We conclude that any eventual measurement of a rate of period change for a given pulsating low-mass pre-white dwarf or white dwarf star could shed light about its evolutionary status. Also, in view of the systematic difficulties in the spectroscopic classification of stars of the ELM Survey, an eventual measurement of $\dot{\Pi}$ could help to confirm that a given pulsating star is an authentic low-mass white dwarf and not a star from another stellar population.

PREPARING FOR ADVANCED LIGO: A STAR–GALAXY SEPARATION CATALOG FOR THE PALOMAR TRANSIENT FACTORY

Abstract The search for fast optical transients, such as the expected electromagnetic counterparts to binary neutron star mergers, is riddled with false positives (FPs) ranging from asteroids to stellar flares. While moving objects are readily rejected via image pairs separated by ∼1 hr, stellar flares represent a challenging foreground, significantly outnumbering rapidly evolving explosions. Identifying stellar sources close to and fainter than the transient detection limit can eliminate these FPs. Here, we present a method to reliably identify stars in deep co-adds of Palomar Transient Factory (PTF) imaging. Our machine-learning methodology utilizes the random forest (RF) algorithm, which is trained using sources with Sloan Digital Sky Survey (SDSS) spectra. When evaluated on an independent test set, the PTF RF model outperforms the SExtractor star classifier by ∼4%. For faint sources ( mag), which dominate the field population, the PTF RF model produces a ∼19% improvement over SExtractor. To avoid false negatives in the PTF transient-candidate stream, we adopt a conservative stellar classification threshold, corresponding to a galaxy misclassification rate of 0.005. Ultimately, ∼ objects are included in our PTF point-source catalog, of which only ∼106 are expected to be galaxies. We demonstrate that the PTF RF catalog reveals transients that otherwise would have been missed. To leverage its superior image quality, we additionally create an SDSS point-source catalog, which is also tuned to have a galaxy misclassification rate of 0.005. These catalogs have been incorporated into the PTF real-time pipelines to automatically reject stellar sources as non-extragalactic transients.

UNTANGLING THE NEAR-IR SPECTRAL FEATURES IN THE PROTOPLANETARY ENVIRONMENT OF KH 15D

ABSTRACT We report on Gemini/GNIRS observations of the binary T Tauri system V582 Mon (KH 15D) at three orbital phases. These spectra allow us to untangle five components of the system: the photosphere and magnetosphere of star B, the jet, scattering properties of the ring material, and excess near-infrared (near-IR) radiation previously attributed to a possible self-luminous planet. We confirm an early-K subgiant classification for star B and show that the magnetospheric He i emission line is variable, possibly indicating increased mass accretion at certain times. As expected, the H2 emission features associated with the inner part of the jet show no variation with orbital phase. We show that the reflectance spectrum for the scattered light has a distinctive blue slope and spectral features consistent with scattering and absorption by a mixture of water and methane ice grains in the 1–50 μm size range. This suggests that the methane frost line is closer than ∼5 au in this system, requiring that the grains be shielded from direct radiation. After correcting for features from the scattered light, jet, magnetosphere, and photosphere, we confirm the presence of leftover near-IR light from an additional source, detectable near minimum brightness. A spectral emission feature matching the model spectrum of a 10 M J, 1 Myr old planet is found in the excess flux, but other expected features from this model are not seen. Our observations, therefore, tentatively support the picture that a luminous planet is present within the system, although they cannot yet be considered definitive.

Challenges in the automated classification of variable stars in large databases

With ever-increasing numbers of astrophysical transient surveys, new facilities and archives of astronomical time series, time domain astronomy is emerging as a mainstream discipline. However, the sheer volume of data alone - hundreds of observations for hundreds of millions of sources – necessitates advanced statistical and machine learning methodologies for scientific discovery: characterization, categorization, and classification. Whilst these techniques are slowly entering the astronomer’s toolkit, their application to astronomical problems is not without its issues. In this paper, we will review some of the challenges posed by trying to identify variable stars in large data collections, including appropriate feature representations, dealing with uncertainties, establishing ground truths, and simple discrete classes.

Machine learning techniques to select variable stars

In order to perform a supervised classification of variable stars, we propose and evaluate a set of six features extracted from the magnitude density of the light curves. They are used to train automatic classification systems using state-of-the-art classifiers implemented in the R statistical computing environment. We find that random forests is the most successful method to select variables.

GRB 090510: A GENUINE SHORT GRB FROM A BINARY NEUTRON STAR COALESCING INTO A KERR–NEWMAN BLACK HOLE

ABSTRACT In a new classification of merging binary neutron stars (NSs) we separate short gamma-ray bursts (GRBs) into two subclasses. The ones with erg coalesce to form a massive NS and are indicated as short gamma-ray flashes (S-GRFs). The hardest, with erg, coalesce to form a black hole (BH) and are indicated as genuine short GRBs (S-GRBs). Within the fireshell model, S-GRBs exhibit three different components: the proper GRB (P-GRB) emission, observed at the transparency of a self-accelerating baryon- plasma; the prompt emission, originating from the interaction of the accelerated baryons with the circumburst medium; and the high-energy (GeV) emission, observed after the P-GRB and indicating the formation of a BH. GRB 090510 gives the first evidence for the formation of a Kerr BH or, possibly, a Kerr–Newman BH. Its P-GRB spectrum can be fitted by a convolution of thermal spectra whose origin can be traced back to an axially symmetric dyadotorus. A large value of the angular momentum of the newborn BH is consistent with the large energetics of this S-GRB, which reach in the 1–10,000 keV range erg and in the 0.1–100 GeV range erg, the most energetic GeV emission ever observed in S-GRBs. The theoretical redshift that we derive from the fireshell theory is consistent with the spectroscopic measurement , showing the self-consistency of the theoretical approach. All S-GRBs exhibit GeV emission, when inside the Fermi-LAT field of view, unlike S-GRFs, which never evidence it. The GeV emission appears to be the discriminant for the formation of a BH in GRBs, confirmed by their observed overall energetics.

Photometry of some neglected bright cataclysmic variables and candidates

As part of an effort to better characterize bright cataclysmic variables (CVs) which have received little attention in the past light curves of four confirmed systems (CZ Aql, BO Cet, V380 Oph and EF Tuc) and one candidate (Lib 3) are analyzed. For none of these stars time resolved photometry has been published previously. While no variability was found in the case of Lib 3, which thus cannot be confirmed as a CV, the light curves of all other targets are dominated by strong flickering. Modulations on hourly time scales superimposed on the flickering can probably be related to orbital variations in BO Cet and V380 Oph, but not in CZ Aql and EF Tuc. Variations on the time scale of 10 min in CZ Aql, while not yet constituting convincing evidence, together with previous suspicions of a magnetically channeled accretion flow may point at an intermediate polar nature of this star. Some properties of the flickering are quantified in an effort to enlarge the data base for future comparative flickering studies in CVs and to refine the classification of the target stars.

Characterisation of red supergiants in theGaiaspectral range

The infrared Calcium Triplet and its nearby spectral region have been used for spectral and luminosity classification of late-type stars, but the samples of cool supergiants (CSGs) used have been very limited (in size, metallicity range, and spectral types covered). The spectral range of the Gaia Radial Velocity Spectrograph (RVS) covers most of this region but does not reach the main TiO bands in this region, whose depths define the M sequence. We study the behaviour of spectral features around the Calcium Triplet and develop effective criteria to identify and classify CSGs, comparing their efficiency with other methods previously proposed. We measure the main spectral features in a large sample (almost 600) of CSGs from three different galaxies, and we analyse their behaviour through a principal component analysis. Using the principal components, we develop an automatised method to differentiate CSGs from other bright late-type stars, and to classify them. The proposed method identifies a high fraction of the supergiants (SGs) in our test sample, which cover a wide metallicity range (SGs from the SMC, the LMC, and the Milky Way) and with spectral types from G0 up to late-M. In addition, it is capable to separate most of the non-SGs in the sample, identifying as SGs only a very small fraction of them. A comparison of this method with other previously proposed shows that it is more efficient and selects less interlopers. A way to automatically assign a spectral type to the SGs is also developed. We apply this study to spectra at the resolution and spectral range of the Gaia RVS, with a similar success rate. The method developed identifies and classifies CSGs in large samples, with high efficiency and low contamination, even in conditions of wide metallicity and spectral-type ranges.

Stellar classification of CoRoT targets

The CoRoT faint stars channel observed about 163 600 targets to detect transiting planetary companions. Because CoRoT targets are faint (11< r <16) and close to the galactic plane, only a small subsample has been observed spectroscopically. We describe the latest classification scheme used to derive the spectral type of CoRoT targets, which is based on broadband multi-colour photometry. We assess the accuracy of this spectral classification for the first time. We find that the classification method performs better for stars that were observed during the mission-dedicated photometric ground-based campaigns.The luminosity class is wrong for less than 7% of the targets. Generally, the effective temperature of stars classified as early type (O, B, and A) is overestimated. Conversely, the temperature of stars classified as later type tends to be underestimated. This is mainly due to the adverse effect of interstellar reddening. We find that the median error on the effective temperature is less than 5% for dwarf stars classified with a spectral later than F0, but it is worse for earlier type stars, with up to 20\% error for A and late-B dwarfs, and up to 70% for early-B and O-type dwarfs. Similar results are found for giants, with a median error that is lower than 7% for G- and later type giants, but greater than 25% for earlier types. Overall, we find an average median absolute temperature difference |Delta Teff| = 533+\-6 K for the whole sample of stars classified as dwarfs and |Delta Teff| = 280+\-3 K for the whole sample of giant stars. The corresponding standard deviation is of about 92+\-5 K for dwarfs and 304+\-4 K for giants. Typically for late-type stars, this means that the classification is accurate to about half a class.