Effectively Updating Co-location Patterns in Evolving Spatial Databases

摘要

Spatial co-location mining has been used for discovering spatial event sets which show frequent association relationships based on the spatial neighborhood. This paper presents a problem of finding co-location patterns on evolving spatial databases which are constantly updated with fresh data. Maintaining discovered spatial patterns is a complicated process when a large spatial database is changed because new data points make spatial relationships with existing data points on the continuous space as well as among themselves. The change of neighbor relations can affect co-location mining results with invalidating existing patterns and introducing new patterns. This paper presents an algorithm for effectively updating co-location analysis results and its experimental evaluation.