DC Field | Value | Language |
---|---|---|
dc.contributor.author | Yoo, Jaemin | ko |
dc.contributor.author | Sael, Lee | ko |
dc.date.accessioned | 2023-08-16T00:00:22Z | - |
dc.date.available | 2023-08-16T00:00:22Z | - |
dc.date.created | 2023-08-15 | - |
dc.date.created | 2023-08-15 | - |
dc.date.issued | 2021-05-13 | - |
dc.identifier.citation | 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2021, pp.143 - 155 | - |
dc.identifier.issn | 2945-9133 | - |
dc.identifier.uri | http://hdl.handle.net/10203/311556 | - |
dc.description.abstract | How can we accurately classify feature-based data such that the learned model and results are more interpretable? Interpretability is beneficial in various perspectives, such as in checking for compliance with exiting knowledge and gaining insights from decision processes. To gain in both accuracy and interpretability, we propose a novel tree-structured classifier called Gaussian Soft Decision Trees (GSDT). GSDT is characterized by multi-branched structures, Gaussian mixture-based decisions, and a hinge loss with path regularization. The three key features make it learn short trees where the weight vector of each node is a prototype for data that mapped to the node. We show that GSDT results in the best average accuracy compared to eight baselines. We also perform an ablation study of the various structures of covariance matrix in the Gaussian mixture nodes in GSDT and demonstrate the interpretability of GSDT in a case study of classification in a breast cancer dataset. | - |
dc.language | English | - |
dc.publisher | Springer Science and Business Media Deutschland GmbH | - |
dc.title | Gaussian Soft Decision Trees for Interpretable Feature-Based Classification | - |
dc.type | Conference | - |
dc.identifier.wosid | 000719261700012 | - |
dc.identifier.scopusid | 2-s2.0-85111090253 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 143 | - |
dc.citation.endingpage | 155 | - |
dc.citation.publicationname | 25th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2021 | - |
dc.identifier.conferencecountry | II | - |
dc.identifier.conferencelocation | Virtual | - |
dc.identifier.doi | 10.1007/978-3-030-75765-6_12 | - |
dc.contributor.localauthor | Yoo, Jaemin | - |
dc.contributor.nonIdAuthor | Sael, Lee | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.