@inproceedings{64f26d924be043b0b4f5947fc87f8951,
title = "Context-sensitive regression analysis for distributed data",
abstract = "A precondition of existing ensemble-based distributed data mining techniques is the assumption that contributing data are identically and independently distributed. However, this assumption is not valid in many virtual organization contexts because contextual heterogeneity exists. Focusing on regression tasks, this paper proposes a context-based meta-learning technique for horizontally partitioned data with contextual heterogeneity. The predictive performance of our new approach and the state of the art techniques are evaluated and compared on both simulated and real-world data sets.",
author = "Yan Xing and Madden, \{Michael G.\} and Jim Duggan and Lyons, \{Gerard J.\}",
year = "2005",
doi = "10.1007/11527503\_35",
language = "English",
isbn = "354027894X",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "292--299",
booktitle = "Advanced Data Mining and Applications - First International Conference, ADMA 2005, Proceedings",
note = "1st International Conference on Advanced Data Mining and Applications, ADMA 2005 ; Conference date: 22-07-2005 Through 24-07-2005",
}