1{2 "record_id": "5bcd886348a12f52aa055354a9647ba5c4fa64c4",3 "eligible": true,4 "short_ref": "Almaas, Oust & Pollestad (2026)",5 "country": "Norway",6 "city": "Oslo",7 "property_type": "condominiums",8 "price_type": "transaction price",9 "n_obs": 15702,10 "sample_period": "January 2022 - December 2023",11 "data_source": "Boligmappa AS transactions matched with 393,914 listing images; crowdsourced condition labels",12 "methods_compared": [13 "hedonic linear regression (HLR)",14 "XGBoost",15 "each without condition / with computer-vision condition / with human-labelled condition"16 ],17 "best_method": "XGBoost with human-labelled condition (XGBoost with computer-vision condition: MAPE 7.71%)",18 "metric": "MAPE",19 "best_value": 7.49,20 "hedonic_value": 12.41,21 "test_set": "holdout",22 "features_ai": "interior-photo CNN",23 "incremental_ai_gain": -0.32,24 "incremental_ai_metric": "MAPE (percentage points, XGBoost)",25 "marginal_effects": "SHAP",26 "notes": "AI feature = CNN classification of listing interior photos (room type + 1-5 condition). Table 4, stratified 70/15/15 split, test set: HLR MAPE 12.75% (no condition) -> 12.42% (CV condition) -> 12.41% (human), R2 0.672 -> 0.687 -> 0.690; XGBoost MAPE 8.03% -> 7.71% -> 7.49%, R2 0.847 -> 0.855 -> 0.863. incremental_ai_gain recorded for XGBoost (8.03 -> 7.71, -0.32 pp); for HLR -0.33 pp. Coefficients of condition in HLR (Tables 5-7) and SHAP dependence plots for XGBoost.",27 "confidence": "high"28}