[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-116954-en":3,"doc-seo-116954-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},116954,34359740700684,"Finn","https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487",8,"Research & Report","Property valuation with interpretable machine learning - Master’s Thesis","Property valuation is a critical task for banks, local authorities, developers, and brokers, yet real estate markets often lack clear market value due to infrequent trades, limited supply, negotiated pricing, and highly specific submarkets. Traditionally valuations rely on expert appraisers, while accurate machine learning can face adoption barriers from limited interpretability. This thesis tests whether interpretable machine learning can be feasible by comparing XGB and EBM, focusing on prediction accuracy and the ability to understand model decisions.","Master’s Programme in Industrial Engineering and Management  \nProperty valuation with interpretable machine learning  \nKaapro Hartikainen  \nMaster’s Thesis 2023  \nCopyright ©2023 Kaapro Hartikainen  \n| Author Kaapro Hartikainen |\n| --- |\n| Title of thesis Property valuation with interpretable machine learning |\n| Programme Industrial Engineering and Management |\n| Major Organisation Design and Leadership |\n| Thesis supervisor Prof. Lauri Saarinen |\n| Thesis advisor(s) PhD. Ruth Kaila |\n| Collaborative partner Nordea |\n| Date 24.02.2023 Number of pages 69 Language English |\n| Abstract\u003Cbr>Property valuation is an important task for various stakeholders, including banks, local authorities, property developers, and brokers. As a result of the characteristics of the real estate market, such as the infrequency of trades, limited supply, negotiated prices, and small submarkets with unique traits, there is no clear market value for properties. Traditionally property valuations are done by expert appraisers. Property valuation can also be done accurately with machine learning methods, but the lack of interpretability with accurate machine learning methods can limit the adoption of those methods. Interpretable machine learning methods could be a solution to this issue, but there are concerns related to the accuracy of these methods.\u003Cbr>This thesis aims to evaluate the feasibility of interpretable machine learning methods in property valuation by comparing a promising interpretable method to a more complex machine learning method that has had good results in property valuation previously. The promising interpretable method and the well-performed machine learning method are chosen based on previous literature.\u003Cbr>The two chosen methods, Extreme Gradient Boosting (XGB) and Explainable Boosting Machine (EBM) are compared in terms of prediction accuracy of properties in six big municipalities of Denmark. In addition to the accuracy comparison, the interpretability of the EBM is highlighted.\u003Cbr>The accuracy of the XGB method is better, even though there are no big differences between the two methods in individual municipalities. The interpretability of the EBM is good, as it is possible to understand, how the model makes predictions in general, and how individual predictions are made. |\n| Keywords Property valuation, machine learning, interpretable machine learning, XGBoost, XGB, Explainable Boosting machine, EBM |\n\n| Tekijä Kaapro Hartikainen |\n| --- |\n| Työn nimi Kiinteistöjen arviointi ymmärrettävällä koneoppimisella |\n| Koulutusohjelma Tuotantotalous |\n| Pääaine Organisaatioiden suunnittelu ja johtaminen |\n| Vastuuopettaja/valvoja Prof. Lauri Saarinen |\n| Työn ohjaaja(t) TkT Ruth Kaila |\n| Yhteistyötaho Nordea |\n| Päivämäärä 24.02.2023 Sivumäärä 69 Kieli Englanti |\n| Tiivistelmä\u003Cbr>Kiinteistöjen arviointi on tärkeä tehtävä eri sidosryhmien, kuten pankkien, kuntien, kiinteistökehittäjien ja välittäjien kannalta. Kiinteistömarkkinoiden ominaisuudet, kuten harvoin tapahtuvat kaupat, rajoitettu tarjonta, neuvotellut hinnat ja paikalliset erot, vaikuttavat siihen, että kiinteistöillä ei ole selkeää markkina-arvoa. Perinteisesti kiinteistöjenarvioinnin tekevät asiantuntijat. Kiinteistöjen arviointi voidaan tehdä tarkasti myös koneoppimismenetelmillä, mutta tulkittavuuden puute tarkoilla koneoppimismenetelmillä voi rajoittaa näiden menetelmienkäyttöönottoa. Tulkittavat koneoppimismenetelmät voisivat olla ratkaisutähän ongelmaan, mutta näiden menetelmien tarkkuus ei välttämättä ole tarvittavalla tasolla.\u003Cbr>Tämän työn tavoitteena on arvioida tulkittavienkoneoppimismenetelmien toteutettavuutta kiinteistöjen arvioinnissavertaamalla lupaavaa tulkittavissa olevaa menetelmää monimutkaisempaan koneoppimismenetelmään, jolla on aiemmin saatu hyviä tuloksiakiinteistöjen arvioinnissa. Lupaava tulkittava menetelmä ja hyvin tuloksia saanut koneoppimismenetelmä valitaan aikaisemman kirjallisuuden perusteella.\u003Cbr>Valittuja menetelmiä, Extreme Gradient Boosting","cbCainebAQMvdh9q","https://ap.wps.com/l/cbCainebAQMvdh9q","pdf",2080871,1,69,"English","en",105,"# Introduction\n## Research problem and research questions\n## Research design and scope\n## Structure of the thesis\n# Interpretability in machine learning\n## Why interpretability?\n## Adoption of machine learning\n## Mismatch of goals\n## Confirming important criteria\n## High stakes\n## Gaining knowledge\n## Troubleshooting and improving through iterations\n## Legal requirements","[{\"question\":\"Why is interpretability important in machine learning for property valuation?\",\"answer\":\"Property valuation involves high-stakes decisions where stakeholders need to understand how predictions are formed. Interpretability helps validate important criteria and support adoption of machine learning models.\"},{\"question\":\"Which interpretable and which more complex methods are compared in the thesis?\",\"answer\":\"The thesis compares Extreme Gradient Boosting (XGB) and Explainable Boosting Machine (EBM), selecting both based on prior literature.\"},{\"question\":\"What are the main results regarding accuracy and interpretability?\",\"answer\":\"XGB achieves better overall prediction accuracy, while differences across individual municipalities are small. EBM provides strong interpretability, allowing both general understanding of predictions and explanation of individual predictions.\"}]","Property valuation with interpretable machine learning - Master’s Thesis | PDF",1785672817,174,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"property-valuation-with-interpretable-machine-learning-masters-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/property-valuation-with-interpretable-machine-learning-masters-thesis/116954/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why is interpretability important in machine learning for property valuation?","Question",{"text":75,"@type":76},"Property valuation involves high-stakes decisions where stakeholders need to understand how predictions are formed. Interpretability helps validate important criteria and support adoption of machine learning models.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which interpretable and which more complex methods are compared in the thesis?",{"text":80,"@type":76},"The thesis compares Extreme Gradient Boosting (XGB) and Explainable Boosting Machine (EBM), selecting both based on prior literature.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the main results regarding accuracy and interpretability?",{"text":84,"@type":76},"XGB achieves better overall prediction accuracy, while differences across individual municipalities are small. EBM provides strong interpretability, allowing both general understanding of predictions and explanation of individual predictions.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]