English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

For supervised classification problems involving design, control, otherpractical purposes, users are not only interested in finding a highly accurateclassifier, but they also demand that the obtained classifier be easilyinterpretable. While the definition of interpretability of a classifier canvary from case to case, here, by a humanly interpretable classifier we restrictit to be expressed in simplistic mathematical terms. As a novel approach, werepresent a classifier as an assembly of simple mathematical rules using anon-linear decision tree (NLDT). Each conditional (non-terminal) node of thetree represents a non-linear mathematical rule (split-rule) involving featuresin order to partition the dataset in the given conditional node into twonon-overlapping subsets. This partitioning is intended to minimize the impurityof the resulting child nodes. By restricting the structure of split-rule ateach conditional node and depth of the decision tree, the interpretability ofthe classifier is assured. The non-linear split-rule at a given conditionalnode is obtained using an evolutionary bilevel optimization algorithm, in whichwhile the upper-level focuses on arriving at an interpretable structure of thesplit-rule, the lower-level achieves the most appropriate weights(coefficients) of individual constituents of the rule to minimize the netimpurity of two resulting child nodes. The performance of the proposedalgorithm is demonstrated on a number of controlled test problems, existingbenchmark problems, and industrial problems. Results on two to 500-featureproblems are encouraging and open up further scopes of applying the proposedapproach to more challenging and complex classification tasks.

Interpretable Rule Discovery Through Bilevel Optimization of Split-Rules of Nonlinear Decision Trees for Classification Problems