1{2 "record_id": "2cdb6f374417590715ab5db24f593d17616b9ea2",3 "eligible": false,4 "exclusion_reason": "No regression coefficients are printed: the paper compares predictive performance of linear regression, random forest, XGBoost and gradient boosting (Table 2: R2, MSE, MAE, MedAE) and interprets the gradient-boost model with SHAP importance plots, partial dependence and interaction figures; the linear-regression coefficients for the structural, walkability, streetscape and locational variables are not reported (only definitions and descriptives in Table 1). SHAP values are not marginal effects usable in the coefficient synthesis.",5 "study": {6 "title": "Effects of neighborhood streetscape on the single-family housing price: Focusing on nonlinear and interaction effects using interpretable machine learning",7 "year": 2025,8 "doi": "10.1371/journal.pone.0323495",9 "country": "South Korea",10 "region_city": "Seoul",11 "sample_start": 2017,12 "sample_end": 2019,13 "property_type": "single_family",14 "price_type": "transaction",15 "data_source": "13,776 single-family house transactions in Seoul 2017-2019; Google Street View panoramas (233,106) segmented with DeepLab V3+",16 "document_type": "journal",17 "estimators": [18 "OLS",19 "ML_marginal"20 ],21 "spatial_method": "none (latitude/longitude as features)",22 "time_fe": false,23 "location_fe": false,24 "se_type": "unknown",25 "n_obs_total": 13776,26 "ai_methods": [27 "computer_vision_streetscape",28 "gradient_boosting",29 "XGBoost",30 "random_forest",31 "SHAP"32 ],33 "notes": "Dependent variable ln(price, KRW). Gradient boost R2 0.817 vs linear regression 0.660 (20% hold-out). Streetscape variables: greenery, enclosure, paved sidewalks (log of segmentation shares). Relevant for the AI/computer-vision synthesis (RQ6) qualitatively only."34 },35 "models": [],36 "coefficients": []37}