Unsupervised feature correlation-based spatial stratification for local context-aware modelling
Context-aware modelling improves the accuracy of spatial inferences through using local environmental conditions, spatial dependency, and heterogeneity. However, traditional context-aware approaches generally require constructing separate models for each location, leading to high computational compl...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | GIScience & Remote Sensing |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/15481603.2025.2539556 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Context-aware modelling improves the accuracy of spatial inferences through using local environmental conditions, spatial dependency, and heterogeneity. However, traditional context-aware approaches generally require constructing separate models for each location, leading to high computational complexity and substantial time and resource demands. This study develops an unsupervised feature correlation-based spatial hierarchical approach for local context-aware modeling (UFCSS-LC), and implement the model in the fusion analysis of climate model precipitation data in Qinghai. The results demonstrate that UFCSS-LC significantly enhances modeling accuracy compared to both single learning models and traditional spatial context-aware models in terms of the decision coefficient and root mean square error. The findings demonstrate the UFCSS-LC’s ability in capturing regional characteristics, constructing locally adaptive models, and balancing model complexity with prediction accuracy. The developed approach provides an efficient and accurate solution for spatial inferences and data analysis. |
|---|---|
| ISSN: | 1548-1603 1943-7226 |