The process of knowledge discovery in databases: a first sketch

摘要: The general idea of discovering knowledge in large amounts data is both appealing and intuitive. Typically we focus our attention on learning algorithms, which provide the core capability generalizing from numbers small, very specific facts to useful high-level rules; these technique's seem hold most excitement perhaps substantive scientific content discovery databases (KDD) enterprise. However, when engage real-world tasks, find that they can be extremely complex, induction rules only one small part overall process. While others have written overviews concept KDD, even provided block diagrams for "knowledge systems," no has begun identify all building blocks a realistic KDD process. This what attempt do here. Besides bringing into discussion several parts process received inadequate community, careful elucidation steps framework comparison different technologies tools are almost impossible compare without clean model.