Dealing with Incomplete Data of Category Attributes
Xin She Yang · Huadong Li-Gong Daxue xuebao · 2003
Incomplete data including noisy and missing data occurs frequently in data mining. In this paper we discuss the use of domain knowledge, such as integrity constraints or concept hierarchy, to reengineer the database and allocate sets to which missing or unacceptable outlying data may belong. Attributeoriented knowledge discovery method is proved to be a powerful approach for mining incomplete data in the large database in our experiment of this paper.