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.
The commented out code at line 44&45 of
moa.classifiers.core.conditionaltests.NominalAttributeMultiwayTest.branchForInstance(Instance) should be put back in. As it is, a wrong split attribute is selected for selecting the branch.
For me, my class index is 32 and the split attribute index should be 33. But it branchForInstance actually returns 32 which is the class index and thus I get perfect accuracy.
Attached training and test data files. 'label' is the class.
sampledata_class_label.zip
The commented out code at line 44&45 of moa.classifiers.core.conditionaltests.NominalAttributeMultiwayTest.branchForInstance(Instance) should be put back in. As it is, a wrong split attribute is selected for selecting the branch. For me, my class index is 32 and the split attribute index should be 33. But it branchForInstance actually returns 32 which is the class index and thus I get perfect accuracy. Attached training and test data files. 'label' is the class. sampledata_class_label.zip