ben/ObsidianVault
1
0 Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

vault backup: 2026-06-08 13:16:43

Browse Source
This commit is contained in:
ben committed 2026-06-08 13:16:43 -07:00
1 parent bf1a9d6c2b
commit 3b8d647184
1 file changed
+1 -1
Download Patch File Download Diff File Expand all files Collapse all files
Wiki/Machine Learning/Imbalanced Data.md
+1 -1
View File
@@ -16,5 +16,5 @@ The synthetic minority oversampling technique (SMOTE) is used to generate synthe
4. Create new synthetic data point along line between first point and selected neighbors 4. Create new synthetic data point along line between first point and selected neighbors
5. Repeat 5. Repeat
![[SMOTE.excalidraw]] ![[SMOTE.excalidraw]]
SMOTE can be helpful to increase the size of a minority dataset but SMOTE can be helpful to increase the size of a minority dataset but has a few limitations. If the minority class is too small SMOTE can overgeneralize. It also does not work very well with categorical data and can create some data points with values that don't make sense for the feature. There is also no way for it to predict/generate outliers which could cause a model to overfit the data.
### Algorithm-Level ### Algorithm-Level