J/MNRAS/491/2280ivo://CDS.VizieR/J/MNRAS/491/2280doi:10.26093/cds/vizier.74912280CDSSharma K.Kembhavi A.Kembhavi A.Sivarani T.Abraham S.Vaghmare K.2024-08-20T20:14:35Z2023-04-13T15:09:27ZCDS support team

CDS, Observatoire de Strasbourg, 11 rue de l'Universite, F-67000 Strasbourg, France

cds-question@unistra.frastronomical-modelsmorgan-keenan-classificationstellar-spectral-typesvisible-astronomyspectroscopyDue to the ever-expanding volume of observed spectroscopic data from surveys such as SDSS and LAMOST, it has become important to apply artificial intelligence (AI) techniques for analysing stellar spectra to solve spectral classification and regression problems like the determination of stellar atmospheric parameters T_eff_, logg, and [Fe/H]. We propose an automated approach for the classification of stellar spectra in the optical region using convolutional neural networks (CNNs). Traditional machine learning (ML) methods with 'shallow' architecture (usually up to two hidden layers) have been trained for these purposes in the past. However, deep learning methods with a larger number of hidden layers allow the use of finer details in the spectrum which results in improved accuracy and better generalization. Studying finer spectral signatures also enables us to determine accurate differential stellar parameters and find rare objects. We examine various machine and deep learning algorithms like artificial neural networks, Random Forest, and CNN to classify stellar spectra using the Jacoby Atlas, ELODIE, and MILES spectral libraries as training samples. We test the performance of the trained networks on the Indo-U.S. Library of Coude Feed Stellar Spectra (CFLIB). We show that using CNNs, we are able to lower the error up to 1.23 spectral subclasses as compared to that of two subclasses achieved in the past studies with ML approach. We further apply the trained model to classify stellar spectra retrieved from the SDSS data base with SNR>20.2020MNRAS.491.2280Shttps://cdsarc.cds.unistra.fr/viz-bin/cat/J/MNRAS/491/2280CatalogResearchIsServedByTAP VizieR generic serviceIsServedByConesearch servicerelated-toIII/92 : A Library of Stellar Spectra (Jacoby+ 1984)J/MNRAS/371/703 : MILES library of empirical spectra (Sanchez-Blazquez+, 2006)J/ApJS/152/251 : Indo-US library of coude feed stellar spectra (Valdes+, 2004)V/154 : Sloan Digital Sky Surveys (SDSS), Release 16 (DR16)https://cds.unistra.fr/vizier-org/licences_vizier.htmlhttps://vizier.cds.unistra.fr/viz-bin/VizieR-2?-source=J/MNRAS/491/2280https://vizier.iucaa.in/viz-bin/VizieR-2?-source=J/MNRAS/491/2280http://vizieridia.saao.ac.za/viz-bin/VizieR-2?-source=J/MNRAS/491/2280https://vizier.cds.unistra.fr/viz-bin/votable?-source=J/MNRAS/491/2280https://vizier.iucaa.in/viz-bin/votable?-source=J/MNRAS/491/2280http://vizieridia.saao.ac.za/viz-bin/votable?-source=J/MNRAS/491/2280GETtext/xml+votablehttps://tapvizier.cds.unistra.fr/TAPVizieR/tapCone search capability for table J/MNRAS/491/2280/table5 (Expected and predicted class for CFLIB spectra)https://vizier.cds.unistra.fr/viz-bin/conesearch/J/MNRAS/491/2280/table5?https://vizier.iucaa.in/viz-bin/conesearch/J/MNRAS/491/2280/table5?http://vizieridia.saao.ac.za/viz-bin/conesearch/J/MNRAS/491/2280/table5?GETtext/xml+votable180.050000true174.170455-9.80222280.0055555555555555566/161 200 311 341 351 377 405 428 433 436 587 684 689 706 737 815 894 915 934 962 975 1040 1082 1218 1247 1249 1286 1297 1360 1397 1399 1408 1425 1535 1556 1603 1696 1788 1808 1827 1873 1885 2133 2169 2234 2317 2433 2440 2460 2472 2476 2489 2525 2549 2619 2637 2650 2664 2771 2828 2834 2837 2923 2970-2971 2997 3020 3026 3036 3077 3079 3085 3087 3118 3138 3229 3242 3390 3430 3572 3600 3619 3639 3662 3741 3777 3849 3878 3913 3975 4038 4041 4146 4191 4206 4355 4373 4419 4428 4438 4444 4468 4506 4512 4559 4574 4808 4846 4854 4890 4895 4905 4929 4972 4990 5009 5059 5073 5109 5149 5154 5230 5239 5245 5266 5299 5304 5323 5339 5349 5522 5530-5531 5535 5587 5631 5694 5718 5776 5802 5805 5833 5836 5929 5982 6039 6051 6056 6066 6130 6153-6154 6165 6183 6203 6210 6219 6255 6298 6308 6326 6339-6340 6381 6539 6671-6672 6705 6730 6740 6761 6766 6776 6783 6792 6825 6827 6882 6897 6926 6941 7016 7039-7040 7101 7113 7115 7129 7131 7136 7141 7174 7177 7195 7221 7238 7252 7373 7384 7390 7402 7413 7486-7487 7560 7610 7710 7958 8007 8009 8109 8198 8278 8314 8333 8345 8354 8372 8439 8496 8498 8523 8671 8689 8716-8717 8746 8767 8773 8781 8790 8792 8822 8843 8845 8923 8985 8995 9004 9061 9197 9213 9303 9310 9327 9373 9382 9403 9491 9502 9507 9537 9610 9642 9702 9753 9792 9813 9835 9873 10067 10101 10130 10161 10164 10194 10266 10302 10359 10375 10380 10405 10458 10476 10595 10598 10632 10661 10696 10731 10757 10784 10800 10803 10812 10817 10834-10835 10871 10886 10899 10926 10937 10997 11109 11150 11177 11193 11197 11247 11250 11314 11339 11370 11441 11492 11547 11599 11620 11644 11671 11691 11773 11795 11851 11858 11885 11907 11927 11930 11935 11959 12030 12056 12082 12271 12295 12332 12424 12451 12472 12512 12516 12621 12848 12865 12882 12965 13025 13031 13033 13040 13075 13079 13087 13268 13292 13478 13620 13688 13721 13777-13778 14001 14027 14109 14206 14302 14400 14458 14484 14615 14625 14631 14641 14653 14753 14763 14802 14810 14850 14862 14878 14901 14911 14922 14927 14931 15019 15021 15030 15045 15082 15088 15096 15144 15170 15174 15226 15237 15273 15288 15297 15311 15320 15356 15363 15368 15396 15481 15519 15719 15722 15747 15768 15940 15986 15994-15995 16023 16035-16036 16144 16213 16834 16883 16892 16895 17012 17047 17061 17124 17158 17171 17245 17256 17266 17391 17394-17395 17425 17432 17445 17538 17551 17573 17576 17578 17588 17606 17608 17768-17769 17860 17919 17941 17955 18236 18276 18289 18304 18315 18416 18553 18561 18591 18615 18797 18990 19017 19110 19179 19189 19202 19309 19531 19547 19585 19676 19745 19825 19921 19923 19985 20015 20110 20137 20278 20308 20313 20322 20390 20440 20446 20847 20886 20921 21112 21117 21196 21233 21269 21285 21307 21317 21324 21389 21397 21457 21467 21515 21612 21647 21696 21699 21710 21723 21738 21833 21862 21864 21879 21901 21993 22118 22168 22180 22205 22250 22263 22275 22317 22398 22419 22518 22553 22558 22594 22654 22708 22765 22818 22822-22823 22868 22900 22933 22949 23007 23011 23015 23096 23140 23161 23190 23201 23279 23323 23369 23375 23408 23421 23466 23525 23533 23538-23540 23545 23547 23593 23611 23637 23675 23695 23717 23765 23776 23796 23820 23822 23864 23887 23938 23974 23979 23983 23996 24028-24029 24048 24062 24085 24103 24151 24164 24212 24336 24370 24374 24382 24411 24420 24480 24536 24570 24921 24924 24951 25051 25198 25246 25278 25304 25410 25435 25444 25487 25567 25607-25608 25659 25741 25754 25769 25776 25789 25793 25869 25910 25930 25941 25950 25963 25973 26014 26024 26110 26260 26311 26320 26347 26366 26457 26475 26486 26521 26544 26552 26575 26589 26595 26657 26676 26758 26783 26975 27014 27033 27040 27082 27106 27198 27255 27322 27330 27574 27581 27694 27719 27752 27757 27763 27793 27858 27975 27979 28003 28019 28044 28103 28146 28149 28151 28157 28173 28215 28336 28342 28350 28352 28394 28436 28439 28469 28471 28480 28486 28494 28498 28528 28538 28660 29052 29077 29095 29136 29347 29445 29521 29589 29621 29680 29765 29776 29783 29791 29893 30106 30114 30225 30227 30285 30304 30307 30328 30339 30398 30439 30442 30448 30482 30501 30510 30576 30640 30643 30655 30693 30709 30712 30732 30798 30818 30953 30965 31061 31073 31078 31091 31094 31116 31121 31160 31206 31390 31411 31416 31428 31478 31556 31600 31614 31683 31686 31698 31751 31756 31759 31768 31776 31842 31930 31956 31979 32070 32087 32095-32096 32102 32109 32127 32140 32142 32170 32182 32185 32286-32287 32291 32357 32364 32433 32468 32472 32477 32485 32522 32542 32604 32610 32612 32619 32621 32651 32661 32688 32757 36564 36651 36664 36693 36697 36759 36762 40408 40635 40639 40716 40722 40777 40795 40805 40882 40885 40932 40943 44378 44439 44485 44669 44722 44747 44834 44847 44860 44879 44883 44903 48534 48884 48967 49008 49049https://cdsarc.cds.unistra.fr/viz-bin/moc/J/MNRAS/491/2280?format=asciiOpticaldefaultJ/MNRAS/491/2280/table5Expected and predicted class for CFLIB spectraSimbadSimbad column added by the CDSmeta.ref.urlcharrecnoRecord number assigned by the VizieR team. Should Not be used for identification.meta.recordintNameStar namemeta.id;meta.maincharExpClassExpected class for CFLIB spectrasrc.spTypecharPreClassPredicted class for CFLIB spectrasrc.spTypecharCFlag[0/1] Consistency between the two spectral classifications (1)meta.codechar_RARight Ascension (J2000) from SIMBAD (not part of the original data)degpos.eq.ra;meta.main_DEDeclination (J2000) from SIMBAD (not part of the original data)degpos.eq.dec;meta.mainJ/MNRAS/491/2280/table8Expected and predicted classes for the SDSS spectraSimbadNameSimbad column added by the CDSmeta.idcharnullableMJDSDSS MJD identifiermeta.idintindexedrecnoRecord number assigned by the VizieR team. Should Not be used for identification.meta.recordintPlateSDSS Plate identifiermeta.id;instr.plateintFiberSDSS Fiber identifiermeta.id;meta.mainintSDSSClassSDSS class for the spectrasrc.spTypecharPreClassPredicted class for the SDSS spectrasrc.spTypecharSpInteractive spectrum from SDSS-DR12meta.ref.urlchar