FastMap를 활용한 효과적 스카이라인 추출

초록

Recently, it becomes more important to extract skyline from the large amounts and high-dimensional data, where the skyline is a set of non-dominated data from which an efficient manipulation for the significant portions of data can be treated with domain characteristics. Conventional research issues have been mainly dedicated to static, numeric, and low dimensional data sets. It is inevitable for the skyline to deal with the data that becomes complex and explosive such as Data Warehousing, CRM, and Business Intelligence, GIS, IR, Patent, Sensor Network, Criminal and Surveillance, Bio and Medical data, Marketing, Culture Industries, etc. In this paper, we solved the curse of dimensionality issue by introducing FastMap and Principal Component Analysis(PCA) method that can reduce high dimensional data to much lower dimensional data without losing solution qualities. We introduced that method to the distance calculation attributes by which more efficient searchable skyline can be derived.