Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Human-Centric Community Detection in Hybrid Metaverse Networks with Integrated AI Entities

Shih-Hsuan Chiu, Ya-Wen Teng, De-Nian Yang, Ming-Syan Chen

TL;DR

This work defines MetaCD, the problem of discovering human-centric communities in hybrid human-AI social networks (HASNs) within Metaverse contexts, and introduces CUSA, a customized AI-aware clustering framework. CUSA combines AI Scoring (via eigenvector, betweenness, and clustering coefficient measures), an AI-aware Louvain modularity optimization, and a 3AC-based adaptive search with probabilistic escaping to balance maximal human closeness with minimal AI presence. The problem is formalized through a human-centric modularity HQ that augments standard modularity with human-to-AI weighting, guiding cluster formation. Four HASN generation strategies are proposed to synthesize HASNs for evaluation, and extensive experiments on real networks (transformed into HASNs) show that CUSA outperforms traditional non-deep and GNN-based baselines in Q and HQ, demonstrating practical viability for human-centric community discovery in mixed human-AI environments. The results highlight the importance of AI-aware design in community detection and offer actionable insights for Metaverse platforms integrating AI companions and agents to foster meaningful human connectivity.

Abstract

Community detection is a cornerstone problem in social network analysis (SNA), aimed at identifying cohesive communities with minimal external links. However, the rise of generative AI and Metaverse introduce complexities by creating hybrid human-AI social networks (denoted by HASNs), where traditional methods fall short, especially in human-centric settings. This paper introduces a novel community detection problem in HASNs (denoted by MetaCD), which seeks to enhance human connectivity within communities while reducing the presence of AI nodes. Effective processing of MetaCD poses challenges due to the delicate trade-off between excluding certain AI nodes and maintaining community structure. To address this, we propose CUSA, an innovative framework incorporating AI-aware clustering techniques that navigate this trade-off by selectively retaining AI nodes that contribute to community integrity. Furthermore, given the scarcity of real-world HASNs, we devise four strategies for synthesizing these networks under various hypothetical scenarios. Empirical evaluations on real social networks, reconfigured as HASNs, demonstrate the effectiveness and practicality of our approach compared to traditional non-deep learning and graph neural network (GNN)-based methods.

Human-Centric Community Detection in Hybrid Metaverse Networks with Integrated AI Entities

TL;DR

This work defines MetaCD, the problem of discovering human-centric communities in hybrid human-AI social networks (HASNs) within Metaverse contexts, and introduces CUSA, a customized AI-aware clustering framework. CUSA combines AI Scoring (via eigenvector, betweenness, and clustering coefficient measures), an AI-aware Louvain modularity optimization, and a 3AC-based adaptive search with probabilistic escaping to balance maximal human closeness with minimal AI presence. The problem is formalized through a human-centric modularity HQ that augments standard modularity with human-to-AI weighting, guiding cluster formation. Four HASN generation strategies are proposed to synthesize HASNs for evaluation, and extensive experiments on real networks (transformed into HASNs) show that CUSA outperforms traditional non-deep and GNN-based baselines in Q and HQ, demonstrating practical viability for human-centric community discovery in mixed human-AI environments. The results highlight the importance of AI-aware design in community detection and offer actionable insights for Metaverse platforms integrating AI companions and agents to foster meaningful human connectivity.

Abstract

Community detection is a cornerstone problem in social network analysis (SNA), aimed at identifying cohesive communities with minimal external links. However, the rise of generative AI and Metaverse introduce complexities by creating hybrid human-AI social networks (denoted by HASNs), where traditional methods fall short, especially in human-centric settings. This paper introduces a novel community detection problem in HASNs (denoted by MetaCD), which seeks to enhance human connectivity within communities while reducing the presence of AI nodes. Effective processing of MetaCD poses challenges due to the delicate trade-off between excluding certain AI nodes and maintaining community structure. To address this, we propose CUSA, an innovative framework incorporating AI-aware clustering techniques that navigate this trade-off by selectively retaining AI nodes that contribute to community integrity. Furthermore, given the scarcity of real-world HASNs, we devise four strategies for synthesizing these networks under various hypothetical scenarios. Empirical evaluations on real social networks, reconfigured as HASNs, demonstrate the effectiveness and practicality of our approach compared to traditional non-deep learning and graph neural network (GNN)-based methods.

Paper Structure

This paper contains 28 sections, 9 equations, 4 figures, 14 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related work
  3. Problem Formulation
  4. Objective Function of MetaCD
  5. Algorithm Design for MetaCD
  6. Empirical Experiments
  7. Experimental Setup
  8. Experimental Results
  9. Comparison of Different AI Node Scoring Methods Used in CUSA
  10. Comparison of CUSA with Baselines on Different Datasets
  11. Comparison of CUSA with Baselines on a Dataset using Different Generation Strategies
  12. Conclusion
  13. Potential Application Scenarios
  14. The Derivation of $\Delta Q$
  15. Computational Complexity of CUSA
  16. ...and 13 more sections

Figures (4)

  • Figure 1: A schematic depiction of HASN, illustrating the interconnected relationships between human users and AI entities. This paper seeks to identify human-centric communities within this hybrid network.
  • Figure 2: Illustrative examples of community detection in an HASN.
  • Figure 3: The framework of the proposed CUSA algorithm.
  • Figure 4: The four proposed hybrid network (HASN) scenarios, where red circles represent AI nodes and other colors represent human nodes.