Research On The Cooperative Technology Of Space - Efficient Co - Location

Over the last decade, spatial database systems are widely and unprecedented used. The development of spatial data collection and storage technique result in the generation of large data in business and scientific field. Due to the growing complexity of the spatial data and its autocorrelation and uncertainty, traditional data analysis techniques exist many limitations, when they are applied to spatial data base. In this case, data mining in spatial database has gradually become a hotspot of research.A spatial co-location pattern is a group of spatial features, whose instances frequently appear in the same region. Mining spatial co-location patterns is one of most important research in spatial association rule mining.In traditional association rule mining, a recent effort has been to consider utility as a new measure of interests, by considering the different values of individual items as utilities.In this paper, we first review previous researches on spatial co-location pattern mining and high utility item mining and illustrate the importance of research of high utility pattern mining in spatial database. Then, we incorporate utility into the spatial pattern mining through the concept of pattern utility, and a general framework for spatial high utility co-location patterns mining is defined.Furthermore, we define the concept of extended pattern utility ratio (EPUR) and partial extended pattern utility ratio (PEPUR). Using their special properties, we further present two pruning algorithms:Extended Pruning Algorithm (EPA) and Partial Pruning Algorithm (PPA) to prune down the number of candidates and can obtain the complete set of spatial high utility co-location patterns. They improve the mining performance and accelerate the spatial high utility co-location pattern generation under different parameter environments.Using synthetic and real-world data sets, substantial experiments show that EPA and PPA effectively and efficiently identify high utility patterns from spatial datasets.