X-TREPAN: a multi class regression and adapted extraction of comprehensible decision tree in artificial neural networks

TL;DR

This paper introduces X-TREPAN, an enhanced algorithm for extracting comprehensible decision trees from neural networks, capable of handling multi-class regression and validated through empirical benchmarks and statistical analysis.

Contribution

The paper extends TREPAN to multi-class regression and adapts it for better interpretability of neural networks, comparing it with C4.5 and validating through experiments.

Findings

X-TREPAN achieves high classification accuracy.

The algorithm effectively handles multi-class regression.

Statistical validation confirms the robustness of results.

Abstract

In this work, the TREPAN algorithm is enhanced and extended for extracting decision trees from neural networks. We empirically evaluated the performance of the algorithm on a set of databases from real world events. This benchmark enhancement was achieved by adapting Single-test TREPAN and C4.5 decision tree induction algorithms to analyze the datasets. The models are then compared with X-TREPAN for comprehensibility and classification accuracy. Furthermore, we validate the experimentations by applying statistical methods. Finally, the modified algorithm is extended to work with multi-class regression problems and the ability to comprehend generalized feed forward networks is achieved.