earticle

영어

A skyline of a n-dimensional data contains the data objects that are not dominated by any other data object on all dimensions. However, as the number of data dimensions increases the probability of domination points become very low, accordingly the number of points in the skyline becomes large. Also skyline search space has been identified as the key problem in real-time multidimensional databases. None of the traditional search techniques include the use of dimensionality reduction to optimize the search space. Skyline query computation on the server consecutively reduces the amount of data transferred between the server sites. Traditional static lower bound and upper bound probability computation will increase the number of non-dominance points. In this proposed work, an optimized skyline boundary detection algorithm is used to filter the skyline objects and pruning the local probability. Also, global probability computation was improved on the large skyline databases in order to minimize the search space and storage .The experimental results show that the efficiency of the proposed approach compared to traditional static skyline bound techniques in terms of time and search space are concerned.