Abstract | ||
---|---|---|
Learning from data with severe class imbalance is difficult. Established solutions include: under-sampling, adjusting classification threshold, and using an ensemble. We examine the performance of combining these solutions to balance the sensitivity and specificity for binary classifications, and to reduce the MSE score for probability estimation. |
Year | Venue | Keywords |
---|---|---|
2011 | Canadian Conference on AI | MSE score,classification threshold,severe class imbalance,imbalanced data,binary classification,established solution,probability estimation |
DocType | Volume | ISSN |
Conference | 6657.0 | 0302-9743 |
Citations | PageRank | References |
7 | 0.52 | 12 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
William Klement | 1 | 21 | 2.90 |
Szymon Wilk | 2 | 461 | 40.94 |
Wojtek Michalowski | 3 | 266 | 41.48 |
Stan Matwin | 4 | 3025 | 344.20 |