Cloud-Based Weighted Kernel Possibilistic c-Means Algorithm for Scalable Big Data Clustering Using AI and Parallel Processing

Abstract : The increasing volume and complexity of big data demand robust, noise-tolerant, and scalable clustering algorithms capable of operating efficiently in distributed computing environments. Traditional clustering approaches often struggle with high-dimensional, noisy data and lack the scalability required for cloud-based applications. This paper introduces a Cloud-Based Weighted Kernel Possibilistic c-Means (CB-WKPCM) algorithm that addresses these limitations by integrating kernel-based learning, weight adaptation, and possibilistic clustering principles within a parallel processing framework.The proposed CB-WKPCM algorithm utilizes artificial intelligence techniques to dynamically optimize kernel parameters and cluster weights, enhancing clustering accuracy and robustness against outliers. Implemented on a distributed cloud infrastructure, the algorithm leverages parallel computing to handle massive datasets with reduced computational overhead. Extensive experimental evaluations on benchmark and real-world datasets demonstrate that CB-WKPCM outperforms traditional clustering algorithms in terms of clustering quality, execution time, and scalability. This research contributes to the field of big data analytics by offering a highly scalable, noise-resilient, and intelligent clustering solution suitable for modern cloud environments.

Keyword Big Data Clustering; Cloud Computing; Weighted Kernel Possibilistic c-Means; Artificial Intelligence; Parallel Processing; Distributed Algorithms; Scalable Data Mining; Noise-Resilient Clustering; Kernel-Based Methods; High-Dimensional Data Analysis.