| Summary: | 碩士 === 國立高雄應用科技大學 === 電子與資訊工程研究所碩士班 === 92 === Spatial data mining is the discovery of interesting relationships and characteristics that may exist implicitly in spatial databases. In this paper, we present three novel algorithms for performing CLARANS clustering and DBSCAN clustering. First, we propose a new clustering method called CRSM, aiming at identifying spatial structures that may be present in the data. Second, building on top of DBSCAN, called CDAP algorithm, we develop a new spatial data mining algorithms aiming at discovering relationships between spatial attributes. This algorithm can discover knowledge that is easy to find with existing spatial data mining algorithms.
Our experimental results demonstrate that our scheme can improve the computational complexity of the CLARANS algorithm based on both the total number of distance calculations and the overall computation time; Especially, the proposed CDAP algorithm, can automatically estimate the two parameters of DBSCAN algorithm, so that improve the clustering performance of DBSCAN algorithm.
|
|---|
