Abstract
Noise handling is an essential task in data mining research and applications. There are three issues in dealing with noisy information sources: noise identification, noise profiling, and noise tolerant mining. During noise identification, erroneous data records are identified and ranked according to their impact or some predefined measures. Class noise and attribute noise can be distinguished at this stage. This identification allows the users to process their noisy data with different priorities based on the data properties. Noise profiling discovers patterns from previously identified errors that can be used to summarize and monitor these data errors. In noise tolerant mining, we integrate the noise profile information into data mining algorithms and boost their performances from the original noisy data. In this talk, I will present our existing and ongoing research efforts on these three issues.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 2007 Springer Berlin Heidelberg
About this paper
Cite this paper
Wu, X. (2007). Class Noise vs Attribute Noise: Their Impacts, Detection and Cleansing. In: Zhou, ZH., Li, H., Yang, Q. (eds) Advances in Knowledge Discovery and Data Mining. PAKDD 2007. Lecture Notes in Computer Science(), vol 4426. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-71701-0_3
Download citation
DOI: https://doi.org/10.1007/978-3-540-71701-0_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-71700-3
Online ISBN: 978-3-540-71701-0
eBook Packages: Computer ScienceComputer Science (R0)