A New Compression Strategy for Spatial Co-Location Patterns Based on Fuzzy β Covering

Spatial co-location patterns (SCPs) are defined as sets of spatial features whose instances are frequently located near each other. While fuzzy spatial structures have advanced SCP mining by modeling real-world spatial uncertainty, they often incur high computational costs and lead to explosive growth in redundant co-locations. Additionally, traditional fuzzy SCP representations lack precision in capturing semantically meaningful spatial relationships, limiting their usefulness for decision support. To address these challenges, this paper proposes a novel fuzzy β-covering-based mining framework, which: (1) employs fuzzy clustering for adaptive spatial soft partitioning to handle uncertainty; (2) introduces a Fuzzy Covering Participation Rate (FCPR) metric to rigorously quantify feature contributions. Comprehensive experiments on two real-world datasets and multiple synthetic benchmarks demonstrate that the proposed approach mitigates pattern explosion by 14.3–76.9%, substantially improving both computational efficiency and analytical utility in uncertainty-aware spatial pattern mining.