loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Paper Unlock

Authors: Jonathan Bodine and Dorit S. Hochbaum

Affiliation: University of California, Berkeley, U.S.A.

Keyword(s): Decision Tree, Principal Component Analysis, Maximum Cut, Classification.

Abstract: Decision trees are a widely used method for classification, both alone and as the building blocks of multiple different ensemble learning methods. The Max-Cut decision tree involves novel modifications to a standard, baseline model of classification decision tree, precisely CART Gini. One modification involves an alternative splitting metric, Maximum Cut, which is based on maximizing the distance between all pairs of observations that belong to separate classes and separate sides of the threshold value. The other modification is to select the decision feature from a linear combination of the input features constructed using Principal Component Analysis (PCA) locally at each node. Our experiments show that this node-based, localized PCA with the novel splitting modification can dramatically improve classification, while also significantly decreasing computational time compared to the baseline decision tree. Moreover, our results are most significant when evaluated on data sets with hi gher dimensions, or more classes. For the example data set CIFAR-100, the modifications enabled a 49% improvement in accuracy, relative to CART Gini, while reducing CPU time by 94% for comparable implementations. These introduced modifications will dramatically advance the capabilities of decision trees for difficult classification tasks. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 3.15.214.185

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Bodine, J. and Hochbaum, D. (2020). The Max-Cut Decision Tree: Improving on the Accuracy and Running Time of Decision Trees. In Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - KDIR; ISBN 978-989-758-474-9; ISSN 2184-3228, SciTePress, pages 59-70. DOI: 10.5220/0010107400590070

@conference{kdir20,
author={Jonathan Bodine. and Dorit S. Hochbaum.},
title={The Max-Cut Decision Tree: Improving on the Accuracy and Running Time of Decision Trees},
booktitle={Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - KDIR},
year={2020},
pages={59-70},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010107400590070},
isbn={978-989-758-474-9},
issn={2184-3228},
}

TY - CONF

JO - Proceedings of the 12th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2020) - KDIR
TI - The Max-Cut Decision Tree: Improving on the Accuracy and Running Time of Decision Trees
SN - 978-989-758-474-9
IS - 2184-3228
AU - Bodine, J.
AU - Hochbaum, D.
PY - 2020
SP - 59
EP - 70
DO - 10.5220/0010107400590070
PB - SciTePress