Wang, X. and Li, F., 2018. Modeling and analysis of buildings' wind resistance and temperature loss process in coastal area. In: Liu, Z.L. and Mi, C. (eds.), Advances in Sustainable Port and Ocean Engineering. Journal of Coastal Research, Special Issue No. 83, pp. 221–224. Coconut Creek (Florida), ISSN 0749-0208. In the analysis process of resisting wind and temperature loss for buildings, there is a problem of poor detection effect by using traditional methods. In order to solve this problem, in this paper, a modeling and analysis method for resisting buildings' wind and temperature loss in coastal area is proposed. Combining the statistical analysis method with the structural stress analysis method, the construction principle of the model is analyzed, and the mechanical property of building to resist wind in coastal areas is researched; according to the map, simple objects and complex objects are managed, so that a complex object oriented conceptual model is designed, and experimental validation is carried out. The experimental results show that the model has a good effect on wind resistance and temperature loss resistance, and is more accurate than the traditional method.

Brownfield-transformed urban parks, particularly those derived from industrial heritage, play a critical role in both cultural preservation and public-space provision. However, existing studies often rely on linear models and general urban contexts, limiting their ability to capture nonlinear, interaction-driven perception and translate analytical results into design-oriented insights. To address this gap, this study develops an interpretable data-driven framework integrating NLP (natural language processing) with explainable machine learning. Using social media reviews from Shougang Park in Beijing, built environmental elements are identified and structured into four dimensions—Accessibility, Safety, Comfort, and Enjoyment. An XGBoost model combined with SHAP analysis is employed to examine variable importance, nonlinear relationships, and interaction effects. The results reveal that visitor satisfaction is governed by heterogeneous and nonlinear relationships rather than independent additive effects. Several variables exhibit threshold-like, diminishing, and inverted-U-shaped patterns, indicating sensitivity to intensity ranges. More importantly, spatial perception emerges from the nonlinear coupling of multiple elements, forming four representative interaction types: compensatory, inverted-U-shaped, context-dependent, and threshold-like relationships. Key interactions are concentrated around industrial landscape, leisure activities, and supporting facilities. Building on these findings, the study translates interactions into design-oriented strategies, emphasizing synergistic configuration, functional balance, moderated development intensity, and context- sensitive programming. By linking interpretable machine learning with spatial design, this research advances an interaction-oriented paradigm and provides a transferable framework for satisfaction-informed evaluation and optimization of brownfields.

Existing industrial heritage research focuses on conservation and adaptive reuse, often relying on case comparisons, policy analysis, with limited attention to how landscape features shape public perception and behavior. Using Beijing Shougang Industrial Heritage Park as a case study, a multi-source framework integrates semantic analysis, surveys, interviews, eye-tracking, and subjective evaluations to quantify the differentiated effects of six landscape element categories on visual attention, interest, and dwell duration. Specifically, two regression models-the "Average Interest Level" model and the "Average Dwell Duration" model-were constructed, and all reported values represent their corresponding standardized coefficients. Facade elements (interest: 0.198; dwell duration: 0.099) and vegetation and greening elements (0.111; 0.199) exhibit positive associations with visual interest and dwell duration, whereas urban furniture (-0.038; -0.218) and paved surfaces (-0.088; -0.098) are negatively associated with these indicators. Way-finding systems show a relatively stronger association with dwell duration (0.124) than with visual interest (0.045). Notably, industrial heritage elements, despite their symbolic and cultural significance, exhibit weak negative associations with visual interest (β = - 0.018) and dwell duration (β = - 0.077). This finding challenges the common assumption that heritage features inherently enhance visitor engagement and suggests that symbolic significance does not necessarily translate into perceptual salience. Their limited perceptual prominence may be associated with material homogeneity, large spatial scale, and low detail legibility. Grounded in the Space-Perception-Behavior (S-P-B) framework, the study identifies three potential explanatory mechanisms linking spatial attributes, perception, and behavioral preferences. The findings contribute to perception-based industrial heritage research by highlighting the mismatch between symbolic value and perceptual engagement and provide empirical support for experience-oriented regeneration and landscape design strategies.