Incremental Mining of Co-locations from Spatial Database

spatial co-locations represent the subsets of spatial features which are frequently located together in geographic space. This paper presents a new problem of finding co-locations on spatial databases which are constantly changed with new data and disappeared data. Discovering co-locations is a complicated process when a large spatial database is changed because new and disappeared data will produce and take away spatial relationships with existing data as well as themselves. The changed relationships will alter the sets of prevalent co-locations with invalidating existing co-locations and producing new co-locations. So efficient incremental mining of co-locations is very indispensable and challenging. This paper presents an algorithm and pruning strategy for efficiently incremental mining on spatial datasets and executes extensive experimental evaluation on "real+synthetic" data sets.