Co-location features extraction for POI under graph database Neo4j
This paper explores the co-occurrence characteristics of urban POI in spatial networks based on graph structure,it breaks through the limitations of traditional Euclidean space.In terms of theory,the spatial distribution correlation characteristics and mining methods under graph structure were sum-marized and compared with Euclidean space,and the co-occurrence pattern judgment method is used to compare and analyze the graph structure with the Euclidean space by analyzing the Apriori algorithm and using the network Voronoi graph to construct a bridge between the connection graph structure and the association rule mining algorithm.At the practical level,Neo4j's advantages in complex associated data processing is brought into full play,and the efficiency of algorithm is improved by using the graph database Neo4j for storage,management,and processing of graph structures.Furthermore,Neo4j is used in this paer for association rule mining,it has been verified that the graph based method is more suitable for urban POI structure features compared with Euclidean space.
spatial data miningcollocated modenetwork Voronoi graphApriori algorithm
万方数据
维普
