Integrating Novel Class Detection with Concept Drifting Data Streams
Most existing data stream classification techniques ignore one important aspect of stream data is the arrival of a novel class. A data stream classification technique that integrates a novel class detection mechanism into classical classifiers, enabling automatic detection of novel classes before the true labels of the novel class instances arrive. The classification model sometimes needs to wait for more test instances to discover similarities among those instances. It shows how to make fast and correct classification decisions under these constraints and apply them to real benchmark data and prove the superiority of the authors' approach.