Dealing with predictive-but-unpredictable attributes in noisy data sources

Authors:
Ying Yang;Xindong Wu;Xingquan Zhu
Affiliations:
University of Vermont, Burlington VT;University of Vermont, Burlington VT;University of Vermont, Burlington VT
Venue:
PKDD '04 Proceedings of the 8th European Conference on Principles and Practice of Knowledge Discovery in Databases
Year:
2004

Citing 0
Cited 10

Data cleaning using belief propagation

Proceedings of the 2nd international workshop on Information quality in information systems
Combining proactive and reactive predictions for data streams

Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Determining noisy instances relative to attributes of interest

Intelligent Data Analysis
The pairwise attribute noise detection algorithm

Knowledge and Information Systems - Special Issue on Mining Low-Quality Data
Identifying noisy features with the Pairwise Attribute Noise Detection Algorithm

Intelligent Data Analysis
The multiple imputation quantitative noise corrector

Intelligent Data Analysis
Conceptual equivalence for contrast mining in classification learning

Data & Knowledge Engineering
Application-Independent Feature Construction from Noisy Samples

PAKDD '09 Proceedings of the 13th Pacific-Asia Conference on Advances in Knowledge Discovery and Data Mining
Improving mining quality by exploiting data dependency

PAKDD'05 Proceedings of the 9th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining
A novel classification algorithm to noise data

ICSI'12 Proceedings of the Third international conference on Advances in Swarm Intelligence - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

Attribute noise can affect classification learning. Previous work in handling attribute noise has focused on those predictable attributes that can be predicted by the class and other attributes. However, attributes can often be predictive but unpredictable. Being predictive, they are essential to classification learning and it is important to handle their noise. Being unpredictable, they require strategies different from those of predictable attributes. This paper presents a study on identifying, cleansing and measuring noise for predictive-but-unpredictable attributes. New strategies are accordingly proposed. Both theoretical analysis and empirical evidence suggest that these strategies are more effective and more efficient than previous alternatives.