An Efficient Hybrid Improved Feature Vector Manifold Clustering with Neighbour Search Optimization

In this paper, the IFMCNSO algorithm a novel hybrid Improved Feature Vector Manifold clustering with Neighbour search optimization clustering algorithm —is presented. Many methods for linear or nonlinear manifold clustering have been developed recently. While in many cases they have proven to perform better than classic clustering algorithms, the majority of these approaches have a high complexity. In order to overcome the clustering problem, particularly for high-dimensional datasets, this work provides an effective hybrid method called IFMCNSO. By using this strategy, the domain in which feature vector manifold learning and Neighbor search optimization techniques can be used is greatly expanded, enabling parameterization in real-world data sets. A good or nearly optimal solution is found using the IFMCNSO algorithm in an acceptable amount of time. A comprehensive comparison of the proposed IFMCNSO algorithm with state-of-the-art clustering algorithms, namely DCNaN, RDMN, HFMST, and HFMST-PSO, reveals that IFMCNSO achieves higher Rand Index (RI) and Adjusted Rand Index (ARI) scores, underscoring its exceptional clustering performance and accuracy