Description

A classification and novel class detection technique for concept-drifting data streams that addresses four major challenges, namely, infinite length, concept-drift, concept-evolution, and feature-evolution. Preprocessing is the integral part of learning process. Data preprocessing is nothing but removing noisy data, missing values, redundant features in the data. If data evolution at different time, the learning model should able to adapt the changes automatically here we discuss about the beneficial to handle adaptivity of preprocessing and adaptivity of the learning model separately. Our previous work addresses the concept-evolution problem in addition to addressing the infinite length and concept-drift problems. Most of the existing data stream classification techniques, including our previous work, assume that the feature space of the data points in the stream is static. This assumption may be impractical for some type of data, for example text data. The system considers the dynamic nature of the feature space and provides an elegant solution for classification and novel class detection when the feature space is dynamic. We show that our approach outperforms state-of-the-art stream classification techniques in classifying and detecting novel classes in real data streams. Data stream classification suffered from the problem of infinite length, concept drift, concept- evolution and feature- evolution in data mining community. Usually data streams are infinite in length and it makes difficult to store and use all the historical data for training.