dc.contributor.author | Hjort, Anders Dahl | |
dc.contributor.author | Scheel, Ida | |
dc.contributor.author | Sommervoll, Dag Einar | |
dc.contributor.author | Pensar, Johan | |
dc.date.accessioned | 2023-11-24T13:36:38Z | |
dc.date.available | 2023-11-24T13:36:38Z | |
dc.date.created | 2023-10-17T18:31:48Z | |
dc.date.issued | 2023 | |
dc.identifier.issn | 0167-9236 | |
dc.identifier.uri | https://hdl.handle.net/11250/3104580 | |
dc.description.abstract | We introduce Locally Interpretable Tree Boosting (LitBoost), a tree boosting model tailored to applications where the data comes from several heterogeneous yet known groups with a limited number of observations per group. LitBoost constraints the complexity of a Gradient Boosted Trees model in a way that allows us to express the final model as a set of local Generalized Additive Models, yielding significant interpretability benefits while still maintaining some of the predictive power of a Gradient Boosted Trees model. We use house price prediction as a motivating example and demonstrate the performance of LitBoost on a data set of observations from different city districts in Oslo (Norway). We also test the robustness of LitBoost in an extensive simulation study on a synthetic data set. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | Elsevier | en_US |
dc.rights | Navngivelse 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by/4.0/deed.no | * |
dc.title | Locally interpretable tree boosting: An application to house price prediction | en_US |
dc.title.alternative | Locally interpretable tree boosting: An application to house price prediction | en_US |
dc.type | Peer reviewed | en_US |
dc.type | Journal article | en_US |
dc.description.version | acceptedVersion | en_US |
dc.source.journal | Decision Support Systems | en_US |
dc.identifier.doi | https://doi.org/10.1016/j.dss.2023.114106 | |
dc.identifier.cristin | 2185677 | |
dc.relation.project | Norges forskningsråd: 322779 | en_US |
dc.relation.project | Norges forskningsråd: 332645 | en_US |
cristin.ispublished | false | |
cristin.fulltext | postprint | |
cristin.qualitycode | 2 | |