Fast Multiview Clustering by Optimal Graph Mining.

Area of Science:

• Machine Learning
• Data Mining
• Computer Science

Background:

• Multiview clustering (MVC) leverages diverse data sources but struggles with large-scale datasets.
• Existing MVC methods often require extensive hyperparameter tuning, limiting practical application.

Purpose of the Study:

• To develop a scalable and efficient multiview clustering algorithm for large datasets.
• To address the limitations of hyperparameter tuning in current MVC approaches.

Main Methods:

• A novel optimal graph mining model is proposed for consistent clustering structure extraction.
• Landmark-based graphs from different views are utilized to identify an optimal graph representation.
• A parameter-free model and an efficient linear-complexity algorithm are developed.

Main Results:

• The proposed model effectively handles large-scale multiview data.
• Experiments demonstrate superior performance compared to existing methods on real-world datasets.
• The parameter-free nature simplifies model application and avoids complex tuning.

Conclusions:

• The developed fast multiview clustering method offers a scalable and efficient solution for large datasets.
• The optimal graph mining approach provides a robust and parameter-free alternative for clustering.
• This work advances the field of large-scale multiview data analysis.