Label Informativeness-Based Minority Oversampling in Graphs (LIMO)

Title

Label Informativeness-Based Minority Oversampling in Graphs (LIMO)

Creator

Das, Rishav; Mohanty, Sikta; Joshi, Rucha Bhalchandra; Mishra, Subhankar

Description

Class imbalance is a pervasive issue in many realworld datasets, particularly in graph-structured data, where certain classes are significantly underrepresented. This imbalance can severely impact the performance of Graph Neural Networks (GNNs), leading to biased learning or over-fitting. The existing oversampling techniques often overlook the intrinsic properties of graphs, such as Label Informativeness (LI), which measures the amount of information a neighbor's label provides about a node's label. To address this, we propose Label Informativenessbased Minority Oversampling (LIMO), a novel algorithm that strategically oversamples minority class nodes by augmenting edges to maximize LI. This technique generates a balanced, synthetic graph that enhances GNN performance without significantly increasing data volume. Our theoretical analysis shows that the effectiveness of GNNs is directly proportional to label informativeness, with mutual information as a mediator. Additionally, we provide insights into how variations in the number of inter-class edges influence the LI by analyzing its derivative. Experimental results on various homophilous and heterophilous benchmark datasets demonstrate the effectiveness of LIMO in improving the performance of node classification for different imbalance ratios, with particularly significant improvements observed in heterophilous graph datasets. Our code is available at https://github.com/smlab-niser/limo. 2025 IEEE.

Source

Proceedings - IEEE International Conference on Knowledge Graph, ICKG 2025;pp.35-42

Date

01-01-2025

Publisher

Institute of Electrical and Electronics Engineers Inc.

Subject

class imbalance; graph neural networks; label informativeness; node classification; oversampling

Coverage

Das R., NISER, OCC of HBNI, Bhubaneswar, India; Mohanty S., Christ University, Bangalore, India; Joshi R.B., The Cyprus Institute, Nicosia, Cyprus; Mishra S., NISER, OCC of HBNI, Bhubaneswar, India

Rights

Restricted Access; Hardcopy may be available in the library

Relation

ISBN: 979-833156689-0;

Format

online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1109/ICKG66886.2025.00012

https://www.scopus.com/pages/publications/105035220010?origin=resultslist