Perturbed Tree


Algorithm Details & Recommendations

  • The Perturbed Tree algorithm is designed to optimize the representation of the probabilistic dependency between a Target variable and the to-be-discretized variable. It is an extension of the Tree discretization algorithm, and it functions as follows:

    • Data Perturbation generates a range of datasets.

    • For each perturbed dataset, a univariate tree is learned to predict the Target variable with the to-be-discretized continuous variable.

    • Extracting the most frequent thresholds produces the final discretization.

  • The Perturbed Tree algorithm takes into account the Minimum Interval Weight and can reduce the number of bins if necessary. It can also be more robust than the simple Tree discretization.

Last updated


Bayesia USA

Bayesia S.A.S.

Bayesia Singapore

Copyright ยฉ 2024 Bayesia S.A.S., Bayesia USA, LLC, and Bayesia Singapore Pte. Ltd. All Rights Reserved.