A Novel Clustering and Matrix Based Computation for Big Data Dimensionality Reduction and Classification

For higher dimensional or "Big Data (BD)" clustering and classification, the dimensions of documents have to be considered. The overhead classifying methods might also reduced by resolving volumetric issue documents. However, shortened collection potentially generate noise abnormalities. Previous abnormality information removal strategies include several different approaches that already been established throughout time. To increase classification accuracy, current classifications new has created conduct must deal with some most difficult issues in BD document categorization clustering. Hence, goals this research are derived from can solved only expanding accuracy classifiers. Superior clusters may achieved using effective "Dimensionality Reduction (DR)". As first step research, we introduce a unique DR approach preserves word frequency collection, allowing algorithm obtain improved (or) at least equal levels lower dimensionality set When "Word Patterns (WPs)" during "WP Clustering (WPC)", imply WP "Similarity Function (SF)" for Computation (SC)" used as part WPC. is accomplished use gained various clusters. Finally, provide Measures" SC high texts deliver SF classification. With assessment criteria like "Information-Ratio Dimension-Reduction", "Accuracy", "Recall", discovered proposed method paired (WP-SC) scaled extremely effectively "Dataset’s (DS)" surpasses technique AFO-MKSVM. According findings, WP-SC produced more favorable outcomes than LDA-SVM AFO-MKSVM approaches.