Incorporating spatial association into statistical classifiers:local pattern-based prior tuning

Bai, H. and Cao, F. and Atkinson, M.P. and Chen, Q. and Wang, J. and Ge, Y. (2020) Incorporating spatial association into statistical classifiers:local pattern-based prior tuning. International Journal of Geographical Information Science, 34 (10). pp. 2077-2114. ISSN 1365-8816

[img]
Text (manuscript)
manuscript.pdf - Accepted Version
Restricted to Repository staff only until 10 March 2021.
Available under License Creative Commons Attribution-NonCommercial.

Download (12MB)

Abstract

This paper proposes a new classification method for spatial data by adjusting prior class probabilities according to local spatial patterns. First, the proposed method uses a classical statistical classifier to model training data. Second, the prior class probabilities are estimated according to the local spatial pattern and the classifier for each unseen object is adapted using the estimated prior probability. Finally, each unseen object is classified using its adapted classifier. Because the new method can be coupled with both generative and discriminant statistical classifiers, it performs generally more accurately than other methods for a variety of different spatial datasets. Experimental results show that this method has a lower prediction error than statistical classifiers that take no spatial information into account. Moreover, in the experiments, the new method also outperforms spatial auto-logistic regression and Markov random field-based methods when an appropriate estimate of local prior class distribution is used.

Item Type:
Journal Article
Journal or Publication Title:
International Journal of Geographical Information Science
Additional Information:
This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Geographical Information Science on 10/03/2020, available online: https://www.tandfonline.com/doi/full/10.1080/13658816.2020.1737702
Uncontrolled Keywords:
/dk/atira/pure/subjectarea/asjc/3300/3305
Subjects:
ID Code:
142739
Deposited By:
Deposited On:
11 May 2020 10:10
Refereed?:
Yes
Published?:
Published
Last Modified:
01 Sep 2020 06:43