SparkSMOTE
The Synthetic Minority Oversampling Technique (SMOTE) implemented in Spark (see original paper). This is a very useful method for dealing with highly imbalanced datasets.
Usage Details
Getting started
Compile and run example data (in data directory). Must specify input and output paths (see algorithmic parameters).
sbt compile
sbt package
./run
Output file will contain the original dataset combined with the artificial instances generated by SMOTE.
Data format
- Any headers must be removed from the data.
- First column corresponds to the datapoint's label (Y). The remaining clumns are the features.
- Labels have to 0 and 1.
- Label 0 (Y=0) corresponds to majority class examples, and label 1 (Y=1) corresponds to minority class examples.
Algorithmic parameters
Parameters that MUST specified in the "run" file:
- inputDirectory: Path to training data file.
- inputDirectory: Path to output of SMOTE.
- numFeatures: Number of features in dataset.
Parameters that can be specified in the "run" file:
- K: Number of nearest neighbors to consider when creating artificial examples. If ommited, the default is K=5.
- oversamplingPctg: Oversampling percentage, i.e. by how much the size of the minority class will be increased. If ommited, the default is oversamplingPctg=1.0 (100%).
- delimiter: Delimiter of dataset. If ommited, the default is delimiter=",".
- numPartitions: After filtering out the majority class data points, the minority class examples are repartitioned -- this prevents ending up with all minority class data points in a single partition and not taking advantage of data parallelism. This variable indicates how many partitions to create. If ommited, the default is numPartitions=20.
