A modified artificial neural network based approach to add new data in a cluster

Clustering is an efficient method to identify natural groupings of data from a large data set to produce a concise representation of a system's behaviour. K-means clustering partitions the observations in given data into K mutually exclusive clusters, and returns a vector of indices indicating to which of the k clusters it has assigned each observation. In k means clustering when a new data is encountered, it gets added to the dataset and is then clustered again which consumes time. The approach presented in this chapter uses artificial neural network trained on previously clustered data to add new data in the corresponding cluster. A new fast converging training algorithm is used for training the network. ANN is known for the accuracy and hence they cluster the new data in exactly similar pattern in which the previous dataset was clustered.