MOA is an open source framework for Big Data stream mining. It includes a collection of machine learning algorithms (classification, regression, clustering, outlier detection, concept drift detection and recommender systems) and tools for evaluation.
I've noticed a few oddities in the multi-label prediction API:
There are multiple MultiLabelInstance classes: moa.core.MultiLabelInstance and com.yahoo.labs.samoa.instances.MultiLabelInstance. Should one of these be removed?
BasicMultiLabelLearner and BasicMultiLabelClassifier are in the moa.classifier.multitarget package, but the moa.classifiers.multilabel package also exists. Should they be moved? Is BasicMultiLabelClassifier even needed?
The classifiers in the moa.classifiers.multilabel package do not implemented the MultiLabelClassifier interface, so they do not show up in the GUI. Are these defunct? Is there some reason they don't implement this interface?
A number of multi-label methods are found in the moa.classifiers.rules.multilabel package, even though they are not rule-based methods. Should all the multi-label methods be moved to moa.classifiers.multilabel?
I've noticed a few oddities in the multi-label prediction API:
MultiLabelInstance
classes:moa.core.MultiLabelInstance
andcom.yahoo.labs.samoa.instances.MultiLabelInstance
. Should one of these be removed?BasicMultiLabelLearner
andBasicMultiLabelClassifier
are in themoa.classifier.multitarget
package, but themoa.classifiers.multilabel
package also exists. Should they be moved? IsBasicMultiLabelClassifier
even needed?moa.classifiers.multilabel
package do not implemented theMultiLabelClassifier
interface, so they do not show up in the GUI. Are these defunct? Is there some reason they don't implement this interface?moa.classifiers.rules.multilabel
package, even though they are not rule-based methods. Should all the multi-label methods be moved tomoa.classifiers.multilabel
?