• Ancy.S
• Cornelius .K

We consider approaches for similarity search in correlated, high- dimensional data sets, which are derived within a clustering framework. Indexing by “vector approximation” (VA-File), which was proposed as a technique to combat the “Curse of Dimensionality,” employs scalar quantization. It ignores dependencies across dimensions, which represents a source of suboptimality. Clustering, on the other hand, exploits interdimensional correlations and is thus a more compact representation of the data set. We developed cluster distance bounds based on separating hyperplane boundaries and our search index, complemented by these bounds, is applicable to Euclidean and Mahalanobis distance metrics. It obtained significant reductions in number of random IOs over several recently proposed indexes, when allowed (roughly) the same number of sequential pages, has a low computational cost and scales well with dimensions and size of the data-set. We note that while the hyperplane bounds are better than MBR and MBS bounds, they are still loose compared with the true query-cluster distance.

R* tree generation, Indexing, hyper plane boundary, prune clusters.

"FILTERING OF HIGH DIMENSIONAL DATA FOR IMPROVING THE SIMILARITY SEARCH CORRELATED DATA USING SPATIAL FILTERING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 6, page no.785-789, June-2018, Available :http://www.jetir.org/papers/JETIRC006139.pdf

"FILTERING OF HIGH DIMENSIONAL DATA FOR IMPROVING THE SIMILARITY SEARCH CORRELATED DATA USING SPATIAL FILTERING", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 6, page no. pp785-789, June-2018, Available at : http://www.jetir.org/papers/JETIRC006139.pdf