Abstract
Big Data analysis is the era of goliath measures of data more than a brief stage like social tagging framework. Social tagging frameworks such as BibSonomy and del.icio.us have turned out to be progressively famous owing to their across-the-board utilization of the web. The social tagging framework could be typical on account of the comments on web 2.0 resources. The social tagging frameworks allow the web clients to clarify distinctive types of web resources with free-form tags. Labels are broadly used to translate and arrange the web 2.0 resources. Tag clustering is characterized as a gathering procedure that implies that the comparable labels are assembled into groups. Tag clustering is exceptionally valuable for sorting out and seeking the web 2.0 resources. Furthermore, it is essential for achieving social tagging frameworks. The objective of feature selection is to decide upon a negligible bookmarked URL subcategory from Web 2.0 data while remembering with reasonably high exactness when speaking to the first bookmarks. In this study, Unsupervised Quick Reduct feature selection calculation is connected so as to locate an arrangement of most frequently tagged bookmarks. Furthermore, clustering techniques such as the Unsupervised Quick Reduct Particle Swarm Optimization (UQRPSO) algorithm are applied for clustering the selected tagged bookmarks, and this algorithm is then compared with k-means clustering (k-means), bat algorithm, and firefly algorithm.