Fig 1.
Schematic diagram illustrating the proposed approach for LBW detection.
Table 1.
Comparison of anomaly detection models.
Table 2.
Categorization of used birth weights datasets.
Table 3.
Comparison of LBW detection results.
Table 4.
Comparison of Very LBW detection results.
Table 5.
Comparison of Extreme LBW detection results.
Table 6.
F-test on the AUCROC scores.
Table 7.
F-test on the AUCPR scores.
Fig 2.
Average rank of features across outlier points for LBW using feature perturbation based interpretation.
Fig 3.
Average rank of features across outlier points for very LBW using feature perturbation based interpretation.
Fig 4.
Average rank of features across outlier points for extreme LBW using feature perturbation based interpretation.