Skip to main content
. 2023 Jun 19;18:4. doi: 10.1186/s13015-023-00228-0

Fig. 1.

Fig. 1

Flow chart of the KBC algorithm. The workflow starts with bootstrap resampling for each base learner. Then, for the neighborhood of each new data point (here, similar patients), weights are assigned to the OOB and resampled sets, and then aggregated into a single score for each learner. Those base learners that surpass the minimum threshold are selected for the prediction of the class label for the new data point