Identifying Group Anchors in Real-World Group Interactions Under Label Scarcity

2025-09-25

AI Summary

This paper addresses the problem of group anchor identification under label-scarce settings—i.e., identifying pivotal individuals in real-world collaborative or communicative groups when only a few groups have known core members (anchors). To this end, we introduce the novel concept of *group anchors* and propose AnchorRadar, a semi-supervised learning framework that jointly models group-level structural patterns and member-level role semantics by integrating graph neural networks with self-supervised signals. Evaluated on 13 real-world datasets, AnchorRadar achieves significantly higher accuracy than state-of-the-art baselines, reduces training time by 10.2×, and requires only 1/43.6 the number of parameters compared to the lightest existing baseline. The method thus demonstrates strong efficacy, computational efficiency, and scalability for large-scale group analysis.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Group interactions occur in various real-world contexts, e.g., co-authorship, email communication, and online Q&A. In each group, there is often a particularly significant member, around whom the group is formed. Examples include the first or last author of a paper, the sender of an email, and the questioner in a Q&A session. In this work, we discuss the existence of such individuals in real-world group interactions. We call such individuals group anchors and study the problem of identifying them. First, we introduce the concept of group anchors and the identification problem. Then, we discuss our observations on group anchors in real-world group interactions. Based on our observations, we develop AnchorRadar, a fast and effective method for group anchor identification under realistic settings with label scarcity, i.e., when only a few groups have known anchors. AnchorRadar is a semi-supervised method using information from groups both with and without known group anchors. Finally, through extensive experiments on thirteen real-world datasets, we demonstrate the empirical superiority of AnchorRadar over various baselines w.r.t. accuracy and efficiency. In most cases, AnchorRadar achieves higher accuracy in group anchor identification than all the baselines, while using 10.2$ imes$ less training time than the fastest baseline and 43.6$ imes$ fewer learnable parameters than the most lightweight baseline on average.

Problem

• Identifying significant individuals called group anchors in real-world group interactions
• Addressing the challenge of group anchor identification under label scarcity conditions
• Developing efficient methods for anchor detection with limited labeled group data

Innovation

• Semi-supervised method for group anchor identification
• Uses information from labeled and unlabeled groups
• Achieves high accuracy with reduced training time