Affiliation:
1. School of Computer Science & Informatics, Cardiff University, Cardiff CF24 4AG, UK
Abstract
Self-supervised representation learning (SSRL) concerns the problem of learning a useful data representation without the requirement for labelled or annotated data. This representation can, in turn, be used to support solutions to downstream machine learning problems. SSRL has been demonstrated to be a useful tool in the field of geographical information science (GIS). In this article, we systematically review the existing research literature in this space to answer the following five research questions. What types of representations were learnt? What SSRL models were used? What downstream problems were the representations used to solve? What machine learning models were used to solve these problems? Finally, does using a learnt representation improve the overall performance?
Subject
Earth and Planetary Sciences (miscellaneous),Computers in Earth Sciences,Geography, Planning and Development
Reference113 articles.
1. Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, MIT Press.
2. Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep Learning, MIT Press.
3. Derrow-Pinion, A., She, J., Wong, D., Lange, O., Hester, T., Perez, L., Nunkesser, M., Lee, S., Guo, X., and Wiltshire, B. (2021, January 1–5). ETA prediction with graph neural networks in Google Maps. Proceedings of the ACM International Conference on Information & Knowledge Management, Online.
4. Joint Deep Learning for land cover and land use classification;Zhang;Remote Sens. Environ.,2019
5. Clinical text data in machine learning: Systematic review;Spasic;JMIR Med. Inform.,2020
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献