Transductive Multiview Modeling With Interpretable Rules, Matrix Factorization, and Cooperative Learning.

Area of Science:

• Artificial Intelligence
• Machine Learning
• Fuzzy Systems

Background:

• Multiview fuzzy systems aim for interpretable models in multiview learning.
• Current methods struggle with efficient collaboration and limited labeled data.

Purpose of the Study:

• To develop a novel transductive multiview fuzzy modeling method.
• To address the challenge of limited labeled data in multiview scenarios.

Main Methods:

• Integrating transductive learning to simultaneously learn models and labels.
• Employing matrix factorization to enhance fuzzy model performance.
• Utilizing collaborative learning across multiple views for robustness.

Main Results:

• The proposed method effectively reduces dependency on labeled data.
• Matrix factorization and collaborative learning improve model performance and robustness.
• Experimental results show competitiveness with existing multiview learning methods.

Conclusions:

• The novel transductive approach offers an effective solution for multiview fuzzy modeling with limited labels.
• The method enhances interpretability and robustness in complex multiview learning tasks.