Similar Prototype Methods for Class Imbalanced Data Classification
2019
In this paper, new methods for solving imbalanced classification problems based on prototypes are proposed. Using similarity relations for the granulation of the universe, similarity classes are generated and a prototype is selected for each similarity class. Experimental results show that the performance of our methods is statistically superior to other imbalanced methods.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
39
References
2
Citations
NaN
KQI