Knowledge Discovery in Heterogeneous Environments
Abstract
This chapter addresses the topic of knowledge discovery in heterogeneous environments. It begins with an overview of the knowledge-discovery process. Because of the importance of using clean, consistent data in the knowledge-discovery process, the chapter focuses on the problems of data integration and cleansing by presenting a framework of semantic conflicts identification and an algorithm for their resolution. The chapter then describes the various data mining tasks that can be performed on the cleansed data, such as association rules, sequential patterns, classification and clustering. It also discusses data mining models and algorithms, such as those related to neural networks, rule induction, decision trees, K-nearest neighbors, and genetic algorithms. The chapter concludes with a summary.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 2002
- Accession Number
- ADA450834
Entities
People
- Magdi N. Kamel
- Marion G. Ceruti
Organizations
- Naval Information Warfare Systems Command