Classification of Missing Geospatial Data from Structure and Mechanism Perspective

Polat Kayali, M. and Basiri, A. (2022) Classification of Missing Geospatial Data from Structure and Mechanism Perspective. In: 30th Annual Geographical Information Science Research UK (GISRUK), Liverpool, UK, 05-08 Apr 2022, (doi: 10.5281/zenodo.6408097)

[img] Text
273653.pdf - Published Version
Available under License Creative Commons Attribution.

303kB

Publisher's URL: https://zenodo.org/record/6408097#.YrMhGHbMLZs

Abstract

Data-centric science, data-empowered society, and policymaking based on data can suffer from flawed conclusions if data are representative, biased, or unavailable. This paper focuses on missingness for which the common mitigation and handling strategies is a deletion or single imputation. However, understanding the reasons causing the missingness can help to understand phenomena better. Distinguishing the different types of missingness help us to develop and implement new imputation approaches, sampling strategies and output uncertainty quantification. In this paper, using missing data mechanism and structure a new taxonomy has been created to classify the causalities of missing geospatial data.

Item Type:Conference Proceedings
Additional Information:The work presented in this paper has been funded by the UK Research and Innovation (UKRI) Future Leaders Fellowship MR/S01795X/2.
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Polat Kayali, Merve and Basiri, Professor Ana
Authors: Polat Kayali, M., and Basiri, A.
College/School:College of Science and Engineering > School of Geographical and Earth Sciences > Earth Sciences
Publisher:Zendoo
Copyright Holders:Copyright © 2022 The Authors
Publisher Policy:Reproduced under a Creative Commons licence

University Staff: Request a correction | Enlighten Editors: Update this record