SX-GeoTree: Self-eXplaining Geospatial Regression Tree Incorporating the Spatial Similarity of Feature Attributions

2025-11-24

AI Summary

Traditional decision trees struggle with modeling spatial dependence and lack local robustness in geographical prediction tasks. To address these issues, this paper proposes the Self-Explaining Geographical Regression Tree (SE-GeoTree). Methodologically, SE-GeoTree embeds spatial similarity into tree construction by jointly optimizing explanation stability and spatial consistency via a consensus similarity network and local Lipschitz continuity constraints. It further introduces a multi-objective splitting criterion integrating geographical weighted regression coefficient distance, SHAP attribution distance, global Moran’s I, and modularity maximization. Evaluated on county-level GDP prediction in Fujian Province and Seattle housing price forecasting, SE-GeoTree achieves predictive accuracy comparable to baseline decision trees (ΔR² < 0.01), while significantly improving residual spatial homogeneity and boosting attribution consensus modularity by 100%.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Ensemble Methods
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Decision trees remain central for tabular prediction but struggle with (i) capturing spatial dependence and (ii) producing locally stable (robust) explanations. We present SX-GeoTree, a self-explaining geospatial regression tree that integrates three coupled objectives during recursive splitting: impurity reduction (MSE), spatial residual control (global Moran's I), and explanation robustness via modularity maximization on a consensus similarity network formed from (a) geographically weighted regression (GWR) coefficient distances (stimulus-response similarity) and (b) SHAP attribution distances (explanatory similarity). We recast local Lipschitz continuity of feature attributions as a network community preservation problem, enabling scalable enforcement of spatially coherent explanations without per-sample neighborhood searches. Experiments on two exemplar tasks (county-level GDP in Fujian, n=83; point-wise housing prices in Seattle, n=21,613) show SX-GeoTree maintains competitive predictive accuracy (within 0.01 $R^{2}$ of decision trees) while improving residual spatial evenness and doubling attribution consensus (modularity: Fujian 0.19 vs 0.09; Seattle 0.10 vs 0.05). Ablation confirms Moran's I and modularity terms are complementary; removing either degrades both spatial residual structure and explanation stability. The framework demonstrates how spatial similarity - extended beyond geometric proximity through GWR-derived local relationships - can be embedded in interpretable models, advancing trustworthy geospatial machine learning and offering a transferable template for domain-aware explainability.

Problem

• Decision trees struggle with capturing spatial dependence in data
• They produce unstable local explanations for geospatial predictions
• Existing methods lack spatially coherent feature attribution robustness

Innovation

• Integrates spatial dependence via Moran's I control
• Enforces explanation robustness through modularity maximization
• Combines GWR coefficients with SHAP attribution distances