Jinghu Xu

An Efficient Approach on Answering Top-k Queries with Grid Dominant Graph Index

The top-k queries are well used in processing spatial data and dimensional data stream to recognize important objects. To address the problem of top-k queries with preferable function, a grid dominant graph (GDG) index based on reserve data points set (RDPS) is presented. Because of the correlation between top-k and skyline computation, when the count of one point’s RDPS is bigger than or equals to the value of k, it is for sure to be pruned the point’s dominant set. This approach in used to prune a lot of free points of top-k queries. GDG used grid index to calculate all RDPS approximately and efficiently. When construction GDG, grid index figures out k-max calculating region to prune free points that all top-k function would not visit, which decreases the amount of index dramatically. When travelling GDG, grid index figures out k-max search region by top-k function which avoids travelling those free points of ad-hoc top-k function. Because GDG uses grid index to rank those data points in the same layer approximately by k-max search region by top-k function, this index structure make less visited points than traditional dominant graph (DG) structure. What’s more, grid index needs less additional computation and storage that make the GDG index more adaptive for top-k queries. Analytical and experimental evidences display that GDG index approach performs better on both storage of index and efficiency of queries.