Community detection with node attributes in multilayer networks.

Area of Science:

• Network Science
• Data Mining
• Machine Learning

Background:

• Community detection is crucial for understanding network structures.
• Existing methods often focus solely on network topology or interactions.
• Multilayer networks involve complex, multi-type interactions.
• Node attributes offer additional valuable information for network analysis.

Purpose of the Study:

• To develop a novel method for community detection in multilayer networks by integrating network topology and node attributes.
• To propose a probabilistic framework that infers attribute-community correlations from data.
• To provide an efficient and scalable algorithmic implementation for practical applications.

Main Methods:

• A principled probabilistic model is proposed to combine multilayer network topology with node attributes.
• The method infers the relationship between attributes and communities without prior assumptions.
• An efficient algorithm is developed, leveraging data sparsity for computational performance.

Main Results:

• The method demonstrates improved performance in community detection compared to attribute-agnostic approaches.
• Incorporating node attributes enhances the prediction of missing links and node attributes.
• The approach yields more interpretable community structures and quantifies attribute impact.

Conclusions:

• Integrating node attributes into multilayer network community detection offers significant advantages.
• The proposed method provides a robust and flexible framework for analyzing complex networks.
• The open-source implementation facilitates broader adoption and further research in network science.