Основные понятия
Community detection in graphs is a challenging problem with various methods explored for accurate results.
Аннотация
This comprehensive review delves into community detection in graphs, categorizing methods into modularity-based, spectral clustering, probabilistic modeling, and deep learning. It introduces the Revised Medoid-Shift method and compares method performance on datasets. The roadmap outlines research background, state-of-the-art methods, research findings, and conclusions. Key methods include Louvain, Infomap, Kernighan-Lin, and spectral clustering. Challenges and advancements in community detection are discussed, emphasizing the importance of accurate and efficient methods.
Статистика
Published as a conference paper at ICLR 2024
Modularity-Based Clustering, Spectral Clustering, Deep Learning, Probabilistic Modeling
Louvain method, Infomap method, Kernighan-Lin method
Цитаты
"The study of complex networks has significantly advanced our understanding of community structures." - Abstract