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.
As of now, Attribute class assumes all values of a nominal attribute are defined before stream processing starts. However, in case stream data evolves, for some attributes the complete list of attribute values may not be known at the beginning. Hence, apart from declaring a priori all attribute values e.g. in ARFF header, it seems reasonable to gradually add new attribute values once they are encountered in the instances. This extension makes it possible.
As of now, Attribute class assumes all values of a nominal attribute are defined before stream processing starts. However, in case stream data evolves, for some attributes the complete list of attribute values may not be known at the beginning. Hence, apart from declaring a priori all attribute values e.g. in ARFF header, it seems reasonable to gradually add new attribute values once they are encountered in the instances. This extension makes it possible.