New RVFL Network Enhances Classification with Fuzzy Logic and Multiview Learning

Random Vector Functional Link (RVFL) networks are known for their rapid training and strong approximation capabilities, but they often struggle with maintaining geometric relationships within data and effectively combining information from multiple feature perspectives. To overcome these limitations, a new model called Intuitionistic Fuzzy Graph Embedded Random Vector Functional Link with Multiview Learning (IFGRVFL-MV) has been introduced. This innovative approach combines three core elements. Firstly, it uses intuitionistic fuzzy sets to manage uncertainty, assigning both membership and non-membership values to data points, which makes the model more resilient to outliers. Secondly, a graph embedding framework is incorporated to preserve the intrinsic topological structures of the data, thereby enhancing the model's ability to generalize. Finally, multiview learning is integrated to leverage complementary information available across different feature spaces. This comprehensive design allows the IFGRVFL-MV model to achieve superior classification accuracy, as demonstrated through experiments on standard benchmark datasets. The results suggest a significant advancement in handling complex data environments characterized by uncertainty and multiple feature views.