English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Forthcoming astronomical surveys are expected to detect new sources in suchlarge numbers that measuring their spectroscopic redshift measurements will benot be practical. Thus, there is much interest in using machine learning toyield the redshift from the photometry of each object. We are particularlyinterested in radio sources (quasars) detected with the Square Kilometre Arrayand have found Deep Learning, trained upon a large optically-selected sample ofquasi-stellar objects, to be effective in the prediction of the redshifts inthree external samples of radio-selected sources. However, the requirement ofnine different magnitudes, from the near-infrared, optical and ultra-violetbands, has the effect of significantly reducing the number of sources for whichredshifts can be predicted. Here we explore the possibility of using machinelearning to impute the missing features. We find that for the training sample,simple imputation is sufficient, particularly replacing the missing magnitudewith the maximum for that band, thus presuming that the non-detection is at thesensitivity limit. For the test samples, however, this does not perform as wellas multivariate imputation, which suggests that many of the missing magnitudesare not limits, but have indeed not been observed. From extensive testing ofthe models, we suggest that the imputation is best restricted to two missingvalues per source. Where the sources overlap on the sky, in the worst case,this increases the fraction of sources for which redshifts can be estimatedfrom 46% to 80%, with >90% being reached for the other samples.

Quasar photometric redshifts from incomplete data using Deep Learning