A transient search using combined human and machine classifications

Darryl E. Wright; Chris J. Lintott; Stephen J. Smartt; Ken W. Smith; Lucy Fortson; Laura Trouille; Campbell R. Allen; Melanie Beck; Mark C. Bouslog; Amy Boyer; K. C. Chambers; Heather Flewelling; Will Granger; Eugene A. Magnier; Adam McMaster; Grant R. M. Miller; James E. O'Donnell; Helen Spiers; John L. Tonry; Marten Veldthuis; Richard J. Wainscoat; Chris Waters; Mark Willman; Zach Wolfenbarger; Dave R. Young

doi:10.1093/mnras/stx1812

A transient search using combined human and machine classifications

Darryl E. Wright, Chris J. Lintott, Stephen J. Smartt, Ken W. Smith, Lucy Fortson, Laura Trouille, Campbell R. Allen, Melanie Beck, Mark C. Bouslog, Amy Boyer, K. C. Chambers, Heather Flewelling, Will Granger, Eugene A. Magnier, Adam McMaster, Grant R. M. Miller, James E. O'Donnell, Helen Spiers, John L. Tonry, Marten VeldthuisRichard J. Wainscoat, Chris Waters, Mark Willman, Zach Wolfenbarger, Dave R. Young

Physics and Astronomy (Twin Cities)

Research output: Contribution to journal › Article › peer-review

34 Scopus citations

Abstract

Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of LSST and other large-throughput surveys.

Original language	Undefined/Unknown
Journal	Monthly Notices of the Royal Astronomical Society
DOIs	https://doi.org/10.1093/mnras/stx1812
State	Published - Jul 17 2017

Bibliographical note

10 pages, 9 figures, submitted to MNRAS

Keywords

astro-ph.IM

Access

10.1093/mnras/stx1812

OpenUrl availability

Full text

Cite this

Wright, D. E., Lintott, C. J., Smartt, S. J., Smith, K. W., Fortson, L., Trouille, L., Allen, C. R., Beck, M., Bouslog, M. C., Boyer, A., Chambers, K. C., Flewelling, H., Granger, W., Magnier, E. A., McMaster, A., Miller, G. R. M., O'Donnell, J. E., Spiers, H., Tonry, J. L., ... Young, D. R. (2017). A transient search using combined human and machine classifications. Monthly Notices of the Royal Astronomical Society. https://doi.org/10.1093/mnras/stx1812

Wright, DE, Lintott, CJ, Smartt, SJ, Smith, KW, Fortson, L, Trouille, L, Allen, CR, Beck, M, Bouslog, MC, Boyer, A, Chambers, KC, Flewelling, H, Granger, W, Magnier, EA, McMaster, A, Miller, GRM, O'Donnell, JE, Spiers, H, Tonry, JL, Veldthuis, M, Wainscoat, RJ, Waters, C, Willman, M, Wolfenbarger, Z & Young, DR 2017, 'A transient search using combined human and machine classifications', Monthly Notices of the Royal Astronomical Society. https://doi.org/10.1093/mnras/stx1812

@article{b6e2f917ca0641ec82b4f88bd5229c06,

title = "A transient search using combined human and machine classifications",

abstract = " Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of LSST and other large-throughput surveys. ",

keywords = "astro-ph.IM",

author = "Wright, {Darryl E.} and Lintott, {Chris J.} and Smartt, {Stephen J.} and Smith, {Ken W.} and Lucy Fortson and Laura Trouille and Allen, {Campbell R.} and Melanie Beck and Bouslog, {Mark C.} and Amy Boyer and Chambers, {K. C.} and Heather Flewelling and Will Granger and Magnier, {Eugene A.} and Adam McMaster and Miller, {Grant R. M.} and O'Donnell, {James E.} and Helen Spiers and Tonry, {John L.} and Marten Veldthuis and Wainscoat, {Richard J.} and Chris Waters and Mark Willman and Zach Wolfenbarger and Young, {Dave R.}",

note = "10 pages, 9 figures, submitted to MNRAS",

year = "2017",

month = jul,

day = "17",

doi = "10.1093/mnras/stx1812",

language = "Undefined/Unknown",

journal = "Monthly Notices of the Royal Astronomical Society",

issn = "0035-8711",

publisher = "Oxford University Press",

}

TY - JOUR

T1 - A transient search using combined human and machine classifications

AU - Wright, Darryl E.

AU - Lintott, Chris J.

AU - Smartt, Stephen J.

AU - Smith, Ken W.

AU - Fortson, Lucy

AU - Trouille, Laura

AU - Allen, Campbell R.

AU - Beck, Melanie

AU - Bouslog, Mark C.

AU - Boyer, Amy

AU - Chambers, K. C.

AU - Flewelling, Heather

AU - Granger, Will

AU - Magnier, Eugene A.

AU - McMaster, Adam

AU - Miller, Grant R. M.

AU - O'Donnell, James E.

AU - Spiers, Helen

AU - Tonry, John L.

AU - Veldthuis, Marten

AU - Wainscoat, Richard J.

AU - Waters, Chris

AU - Willman, Mark

AU - Wolfenbarger, Zach

AU - Young, Dave R.

N1 - 10 pages, 9 figures, submitted to MNRAS

PY - 2017/7/17

Y1 - 2017/7/17

N2 - Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of LSST and other large-throughput surveys.

AB - Large modern surveys require efficient review of data in order to find transient sources such as supernovae, and to distinguish such sources from artefacts and noise. Much effort has been put into the development of automatic algorithms, but surveys still rely on human review of targets. This paper presents an integrated system for the identification of supernovae in data from Pan-STARRS1, combining classifications from volunteers participating in a citizen science project with those from a convolutional neural network. The unique aspect of this work is the deployment, in combination, of both human and machine classifications for near real-time discovery in an astronomical project. We show that the combination of the two methods outperforms either one used individually. This result has important implications for the future development of transient searches, especially in the era of LSST and other large-throughput surveys.

KW - astro-ph.IM

U2 - 10.1093/mnras/stx1812

DO - 10.1093/mnras/stx1812

M3 - Article

SN - 0035-8711

JO - Monthly Notices of the Royal Astronomical Society

JF - Monthly Notices of the Royal Astronomical Society

ER -