High-dimensional data is enlightened by means of massive size of structures such as attributes, make known to introduce new problems to perform clustering. The stretch „Curse of dimensionality', generates a greater number of dimensions that are primarily to explicate the communal rise in time complexity with numerous computational problems. So, the acts of the complete clustering procedures are useless. Therefore, quite a few works have been attentive on presenting new methods are intended for handling higher dimensionality data. But still, the current clustering algorithms facing some open research problems. In this review work, we deliver an instantaneous of the outcome of high-dimensional data space and their consequences for various clustering approaches. Likewise, it presents a detailed overview on comparison of many clustering Ensemble methods for High Dimensional Data such as: Incremental Soft Subspace Semi-Supervised Ensemble Clustering (IS4EC) and Incremental Support Vector Semi-Supervised Subspace Clustering Ensemble (ISVS3CE), which uses the benefits of Random subspace technique, Constraint Propagation approach and Normalized cut algorithm with classifiers to reduce high dimensionality data problem. All these frameworks perform well on datasets with high dimensionality, and results are better than the traditional clustering ensemble methods. The possibility of the forthcoming effort to elaborate the present clustering methods and procedures are discussed at end of the work.
Volume 12 | Issue 4