20.500.12556/RUL-171777 Analysis of apartment prices in Ljubljana’s post-war housing estates (1947–1986) Analiza cen stanovanj v ljubljanskih povojnih stanovanjskih soseskah (1947–1986) This study examines the determinants of apartment prices in 17 post-WWII multi-family housing estates in Ljubljana, Slovenia, constructed between 1947 and 1986. Using 1973 verified transactions from 2020 to 2025, the analysis evaluates spatial, structural, environmental, and accessibility-related variables through a combination of statistical and machine learning techniques. A hedonic price model based on ordinary least squares (OLS) demonstrates modest explanatory power (R2 = 0.171), identifying local market reference prices, floor level, noise exposure, and window renovation as significant predictors. In contrast, seven machine learning models—Random Forest, XGBoost, and Gradient Boosting Machines (GBMs), including optimized versions—achieve notably higher predictive accuracy. The best-performing model, GBM with Randomized Search CV, explains 59.6% of price variability (R2 = 0.5957), with minimal prediction error (MAE = 0.03). Feature importance ana lysis confirms the dominant role of localized price references and structural indicators, while environmental and accessibility variables contribute variably. In addition, three clustering methods (Ward, k-means, and HDBSCAN) are employed to identify typological groups of neighborhoods. While Ward’s and k-means methods consistently identify four robust clusters, HDBSCAN captures greater internal heterogeneity, suggesting five distinct groups and detecting outlier neighborhoods. The integrated approach enhances understanding of spatial housing price dynamics and supports data-driven valuation, urban policy, and regeneration strategies for post-WWII housing estates in Central and Eastern European contexts. Ta študija preučuje dejavnike, ki določajo cene stanovanj v 17 povojnih večstanovanjskih soseskah v Ljubljani, zgrajenih med letoma 1947 in 1986. Na podlagi 1973 preverjenih transakcij iz obdobja 2020–2025 analiza ocenjuje prostorske, strukturne, okoljske in dostopnostne spremenljivke s kombinacijo statističnih in metod strojnega učenja. Hedonični cenovni model na osnovi metode navadnih najmanjših kvadratov (OLS) izkazuje zmerno pojasnjevalno moč (R² = 0,171) ter kot pomembne napovednike identificira lokalne referenčne cene na trgu, nadstropje stanovanja, izpostavljenost hrupu in prenovo oken. Nasprotno pa sedem modelov strojnega učenja – Random Forest (naključni gozd), XGBoost in Gradient Boosting Machines (GBM), vključno z optimiziranimi različicami – doseže bistveno višjo napovedno natančnost. Najuspešnejši model, GBM z naključnim iskanjem parametrov (Randomized Search CV), pojasni 59,6 % variabilnosti cen (R² = 0,5957) ob m inimalni napovedni napaki (MAE = 0,03). Analiza pomembnosti spremenljivk potrjuje prevladujočo vlogo lokalnih referenčnih cen in strukturnih kazalnikov, medtem ko okoljske in dostopnostne spremenljivke prispevajo različno. Poleg tega so bile uporabljene tri metode razvrščanja v skupine (klastriranja) – Ward, k-means in HDBSCAN – za identifikacijo tipoloških skupin sosesk. Metodi Ward in k-means dosledno identificirata štiri robustne grozde, medtem ko HDBSCAN zajame večjo notranjo heterogenost, saj nakazuje pet ločenih skupin in zazna odstopajoče soseske. Integrirani pristop poglablja razumevanje prostorske dinamike cen stanovanj ter podpira podatkovno podprto vrednotenje, urbano politiko in strategije prenove za povojne stanovanjske soseske v srednje- in vzhodnoevropskem kontekstu. post-WWII housing estates real estate market apartment prices random forest XGBoost gradient boosting machines clustering povojne stanovanjske soseske trg nepremičn cene stanovanj razvrščanje v skupine true false true Angleški jezik Slovenski jezik Članek v reviji 2025-09-02 12:40:51 2025-09-02 12:41:00 2025-11-12 11:50:41 0000-00-00 00:00:00 2025 0 0 22 str. iss. 9, art. 1707 Vol. 14 Aug. 2025 0000-00-00 PostprintKoncna Objavljeno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 332.2(497.1)(497.451.1) 2073-445X 10.3390/land14091707 247370755 land-14-01707-v2_(1).pdf land-14-01707-v2_(1).pdf 1 A885EA615E716B94DF3326C15CAB6E00 c81c0e5d8a86c5d95cab34494b5204a91652676187e05f3efc4d2a11ee831699 42616069-87e9-11f0-9328-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=215238 https://www.mdpi.com/2073-445X/14/9/1707 1 2d2b03eb-87e9-11f0-9328-0050569b8976 https://repozitorij.uni-lj.si/Dokument.php?lang=slv&id=215237 Fakulteta za gradbeništvo in geodezijo 0 0 0