Matching Catalogues by Probabilistic Pattern Classification

Abstract

We consider the statistical problem of catalogue matching from a machine learning perspective with the goal of producing probabilistic outputs, and using all available information. A framework is provided that unifies two existing approaches to producing probabilistic outputs in the literature, one based on combining distribution estimates and the other based on combining probabilistic classifiers. We apply both of these to the problem of matching the HIPASS radio catalogue with large positional uncertainties to the much denser SuperCOSMOS catalogue with much smaller positional uncertainties. We demonstrate the utility of probabilistic outputs by a controllable completeness and efficiency trade-off and by identifying objects that have high probability of being rare. Finally, possible biasing effects in the output of these classifiers are also highlighted and discussed.

16 pages, 9 (reduced quality) figures. MNRAS (in press) 2006

References in corpus (6)

• The SuperCOSMOS Sky Survey. Paper I: Introduction and Description
• The HIPASS Catalogue - I. Data Presentation
• The HIPASS Catalogue: III - Optical Counterparts & Isolated Dark Galaxies
• Observations of the Hubble Deep Field with the Infrared Space Observatory. IV. Association of sources with Hubble Deep Field Galaxies
• Exploration of Parameter Spaces in a Virtual Observatory
• Applying Machine Learning to Catalogue Matching in Astrophysics

Cited by in corpus (7)

• Data Mining and Machine Learning in Astronomy
• Probabilistic Cross-Identification of Astronomical Sources
• Evidence for HI replenishment in massive galaxies through gas accretion from the cosmic web
• Bayesian Matching for X-ray and Infrared Sources in the MYStIX Project
• Scientific Data Mining in Astronomy
• Probabilistic positional association of catalogs of astrophysical sources: the Aspects code
• Radio Astronomy in LSST Era