KEPES, Erik, Jakub VRABEL, Ondřej ADAMOVSKÝ, Sara STRITEZSKA, Pavlina MODLITBOVA, Pavel PORIZKA a Jozef KAISER. Interpreting support vector machines applied in laser-induced breakdown spectroscopy. Analytica Chimica Acta. Amsterdam: Elsevier Science publishers, 2022, roč. 1192, February 2022, s. 1-12. ISSN 0003-2670. Dostupné z: https://dx.doi.org/10.1016/j.aca.2021.339352. |
Další formáty:
BibTeX
LaTeX
RIS
@article{2219859, author = {Kepes, Erik and Vrabel, Jakub and Adamovský, Ondřej and Stritezska, Sara and Modlitbova, Pavlina and Porizka, Pavel and Kaiser, Jozef}, article_location = {Amsterdam}, article_number = {February 2022}, doi = {http://dx.doi.org/10.1016/j.aca.2021.339352}, keywords = {LIBS; Classification; Feature importance; SVM; Interpretable machine learning}, language = {eng}, issn = {0003-2670}, journal = {Analytica Chimica Acta}, title = {Interpreting support vector machines applied in laser-induced breakdown spectroscopy}, url = {https://www.sciencedirect.com/science/article/pii/S0003267021011788?via%3Dihub}, volume = {1192}, year = {2022} }
TY - JOUR ID - 2219859 AU - Kepes, Erik - Vrabel, Jakub - Adamovský, Ondřej - Stritezska, Sara - Modlitbova, Pavlina - Porizka, Pavel - Kaiser, Jozef PY - 2022 TI - Interpreting support vector machines applied in laser-induced breakdown spectroscopy JF - Analytica Chimica Acta VL - 1192 IS - February 2022 SP - 1-12 EP - 1-12 PB - Elsevier Science publishers SN - 00032670 KW - LIBS KW - Classification KW - Feature importance KW - SVM KW - Interpretable machine learning UR - https://www.sciencedirect.com/science/article/pii/S0003267021011788?via%3Dihub N2 - Laser-induced breakdown spectroscopy is often combined with a multivariate black box model-such as support vector machines (SVMs)-to obtain desirable quantitative or qualitative results. This approach carries obvious risks when practiced in high-stakes applications. Moreover, the lack of understanding of a black-box model limits the user's ability to fine-tune the model. Thus, here we present four approaches to interpret SVMs through investigating which features the models consider important in the classification task of 19 algal and cyanobacterial species. The four feature importance metrics are compared with popular approaches to feature selection for optimal SVM performance. We report that the distinct feature importance metrics yield complementary and often comparable information. In addition, we identify our SVM model's bias towards features with a large variance, even though these features exhibit a significant overlap between classes. We also show that the linear and radial basis kernel SVMs weight the same features to the same degree. ER -
KEPES, Erik, Jakub VRABEL, Ondřej ADAMOVSKÝ, Sara STRITEZSKA, Pavlina MODLITBOVA, Pavel PORIZKA a Jozef KAISER. Interpreting support vector machines applied in laser-induced breakdown spectroscopy. \textit{Analytica Chimica Acta}. Amsterdam: Elsevier Science publishers, 2022, roč.~1192, February 2022, s.~1-12. ISSN~0003-2670. Dostupné z: https://dx.doi.org/10.1016/j.aca.2021.339352.
|