… Therefore, it is necessary to integrate and evaluate these algorithms to provide the most
recent research trends on data stream classification with novel class detection, which is one of …

… with the novel class instances. Most other data stream classification methods fail to detect
novel classes because they assume that the number of classes in the data stream is fixed. …

… models to compare performance on our streaming datasets. In the case of traditional data
stream classification model with novel class detection, we use the following: (1) One-class …

… class detection problem for data stream classification. Our method differs from the existing
novel class detection … a multi-class model of existing classes while being able to detect the …

… the state-of-the-art classification and novel class detection techniques. … data stream
classification and novel class detection. Section 3 briefly discusses the existing novel class detection …

… possess the potential of forming a new class by showing strong cohesion among themselves…
” classification model and a novel class detection model. Traditional stream classification …

… an elegant solution for classification and novel class detection when the feature space is …
stream classification techniques in classifying and detecting novel classes in real data streams…

… class detector is introduced based on a local density perspective, which allows handling the
data streams with complex class … good classification and novel class detection performance …

… We compare classification and novel class detection performance of SAND with ECSMiner
(Masud et al. 2011). We have chosen ECSMiner since it is a robust and efficient framework …

… stream classification and novelty detection. There have been many works in stream data
classification… approaches - single model classification, and ensemble classification. Some single…