AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Addressing Imbalance in Multi-Label Classification Using Structured Hellinger Forests

Zachary Daniels, Dimitris Metaxas

AAAI 2017 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 53 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
replace axis-parallel info-gain splits with sparse oblique Hellinger loss (ablation)structured forest (SF) · Medicalmacro-F-measure: 0.454 -> 0.775Improved
replace axis-parallel info-gain splits with sparse oblique Hellinger loss (ablation)structured forest (SF) · Mediamillmacro-F-measure: 0.343 -> 0.496Improved
use logistic regression oblique splits instead of axis-parallel splits (ablation)SF · multiple datasetsmacro-F-measure and macro-AUC: Med 0.454 -> 0.685; Enron 0.324 -> 0.252Mixed
add cost-sensitive clustering (ablation)SF-LR · multiple datasetsmacro-F-measure: Med 0.685 -> 0.727; Enron 0.252 -> 0.247Mixed
add cost-sensitive clustering (ablation)SF-H · multiple datasetsmacro-F-measure: Corel 0.231 -> 0.246; Med 0.775 -> 0.792Improved
add cost-sensitive clustering (ablation)SF-H · multiple datasetsmacro-AUC: TMC 0.945 -> 0.946; Yeast 0.708 -> 0.706Mixed
replace logistic regression splits with sparse Hellinger loss (ablation)SF-LR · Medicalmacro-F-measure: 0.685 -> 0.775Improved
compare SOSHF to COCOA and binary relevance random forestSOSHF · 10 benchmarksmacro-F-measure mean rank: mean rank 1.8 vs 5.6 (COCOA), 3.2 (rf-cost)No clear effect

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy