Knowledge Discovery from Geographical Data

摘要: During the last decade, data miners became aware of geographical data. Today, knowledge discovery from geographic is still an open research field but promises to be a solid starting point for developing solutions mining spatiotemporal patterns in knowledge-rich territory. As many concepts feature extraction and are not commonly known within community, need understood before advancing mining, this chapter provides introduction basic In performing spatial set, first important question how use dimension process. At least two viewpoints can considered: either relationships made explicit prior or specialised algorithms directly applied nonspatial The approach claims that somewhat more than other features, and, then, it used prepare set successive step. exploitation selecting values attributes step quite complex, may depend both on structure domain kind model one looking for. This offers advantage allowing reuse standard technology extracted according dimension. second aims at exploiting features dynamically during implies complete reinvention technology, allows flexible knowledge. Mining poses additional challenges, which include background as well handling autocorrelation