Research of mining algorithms for uncertain spatio-temporal co-occurrence pattern

Spatial Co-location pattern mining and Spatio-temporal Co-occurrence pattern mining are important directions of spatial data mining. However, the existing relevant mining algorithms are computational expensive and the algorithms can't effectively deal with uncertain data which are wide spread in many areas. The fast co-occurrence data mining algorithms for the uncertain data are proposed by using the filter-refine method and efficient pruning strategy. The correctness, completeness and complexity of the proposed algorithms are analyzed, and the experimental data shows that the algorithms are effective and reasonable.