Matches in SemOpenAlex for { <https://semopenalex.org/work/W4281680478> ?p ?o ?g. }
Showing items 1 to 68 of
68
with 100 items per page.
- W4281680478 abstract "Objective: Shapley additive explanations (SHAP) is a popular post-hoc technique for explaining black box models. While the impact of data imbalance on predictive models has been extensively studied, it remains largely unknown with respect to SHAP-based model explanations. This study sought to investigate the effects of data imbalance on SHAP explanations for deep learning models, and to propose a strategy to mitigate these effects. Materials and Methods: We propose to adjust class distributions in the background and explanation data in SHAP when explaining black box models. Our data balancing strategy is to compose background data and explanation data with an equal distribution of classes. To evaluate the effects of data adjustment on model explanation, we propose to use the beeswarm plot as a qualitative tool to identify abnormal explanation artifacts, and quantitatively test the consistency between variable importance and prediction power. We demonstrated our proposed approach in an empirical study that predicted inpatient mortality using the Medical Information Mart for Intensive Care (MIMIC-III) data and a multilayer perceptron. Results: Using the data balancing strategy would allow us to reduce the number of the artifacts in the beeswarm plot, thus mitigating the negative effects of data imbalance. Additionally, with the balancing strategy, the top-ranked variables from the corresponding importance ranking demonstrated improved discrimination power. Discussion and Conclusion: Our findings suggest that balanced background and explanation data could help reduce the noise in explanation results induced by skewed data distribution and improve the reliability of variable importance ranking. Furthermore, these balancing procedures improve the potential of SHAP in identifying patients with abnormal characteristics in clinical applications." @default.
- W4281680478 created "2022-06-13" @default.
- W4281680478 creator A5016933525 @default.
- W4281680478 creator A5030967805 @default.
- W4281680478 creator A5040839683 @default.
- W4281680478 creator A5087685316 @default.
- W4281680478 creator A5091448543 @default.
- W4281680478 date "2022-06-08" @default.
- W4281680478 modified "2023-09-27" @default.
- W4281680478 title "Balanced background and explanation data are needed in explaining deep learning models with SHAP: An empirical study on clinical decision making" @default.
- W4281680478 doi "https://doi.org/10.48550/arxiv.2206.04050" @default.
- W4281680478 hasPublicationYear "2022" @default.
- W4281680478 type Work @default.
- W4281680478 citedByCount "0" @default.
- W4281680478 crossrefType "posted-content" @default.
- W4281680478 hasAuthorship W4281680478A5016933525 @default.
- W4281680478 hasAuthorship W4281680478A5030967805 @default.
- W4281680478 hasAuthorship W4281680478A5040839683 @default.
- W4281680478 hasAuthorship W4281680478A5087685316 @default.
- W4281680478 hasAuthorship W4281680478A5091448543 @default.
- W4281680478 hasBestOaLocation W42816804781 @default.
- W4281680478 hasConcept C105795698 @default.
- W4281680478 hasConcept C111472728 @default.
- W4281680478 hasConcept C119857082 @default.
- W4281680478 hasConcept C120936955 @default.
- W4281680478 hasConcept C134306372 @default.
- W4281680478 hasConcept C138885662 @default.
- W4281680478 hasConcept C149782125 @default.
- W4281680478 hasConcept C154945302 @default.
- W4281680478 hasConcept C182365436 @default.
- W4281680478 hasConcept C189430467 @default.
- W4281680478 hasConcept C2776436953 @default.
- W4281680478 hasConcept C2778136018 @default.
- W4281680478 hasConcept C33923547 @default.
- W4281680478 hasConcept C41008148 @default.
- W4281680478 hasConcept C45804977 @default.
- W4281680478 hasConceptScore W4281680478C105795698 @default.
- W4281680478 hasConceptScore W4281680478C111472728 @default.
- W4281680478 hasConceptScore W4281680478C119857082 @default.
- W4281680478 hasConceptScore W4281680478C120936955 @default.
- W4281680478 hasConceptScore W4281680478C134306372 @default.
- W4281680478 hasConceptScore W4281680478C138885662 @default.
- W4281680478 hasConceptScore W4281680478C149782125 @default.
- W4281680478 hasConceptScore W4281680478C154945302 @default.
- W4281680478 hasConceptScore W4281680478C182365436 @default.
- W4281680478 hasConceptScore W4281680478C189430467 @default.
- W4281680478 hasConceptScore W4281680478C2776436953 @default.
- W4281680478 hasConceptScore W4281680478C2778136018 @default.
- W4281680478 hasConceptScore W4281680478C33923547 @default.
- W4281680478 hasConceptScore W4281680478C41008148 @default.
- W4281680478 hasConceptScore W4281680478C45804977 @default.
- W4281680478 hasLocation W42816804781 @default.
- W4281680478 hasLocation W42816804782 @default.
- W4281680478 hasOpenAccess W4281680478 @default.
- W4281680478 hasPrimaryLocation W42816804781 @default.
- W4281680478 hasRelatedWork W1489267121 @default.
- W4281680478 hasRelatedWork W2884834684 @default.
- W4281680478 hasRelatedWork W3088815947 @default.
- W4281680478 hasRelatedWork W3133294580 @default.
- W4281680478 hasRelatedWork W3150895494 @default.
- W4281680478 hasRelatedWork W3160244858 @default.
- W4281680478 hasRelatedWork W3176827378 @default.
- W4281680478 hasRelatedWork W3206132545 @default.
- W4281680478 hasRelatedWork W3213457837 @default.
- W4281680478 hasRelatedWork W4287660321 @default.
- W4281680478 isParatext "false" @default.
- W4281680478 isRetracted "false" @default.
- W4281680478 workType "article" @default.