Text classification using KM-ELM classifier

Classification systems adapts many machine learning techniques for quality performance in data classification. The neural networks has some unique characteristics and features which can handle high dimensional features and documents with noise and contradictory data. Classification is important to classify the input text into different domains appropriately. This paper give out a move towards classification of text that combines two machine learning techniques, K-Means and extreme learning machines. First the clustering and feature selection will perform using K-Means algorithm and then this attribute will be the training set for the extreme learning machine. Extreme learning machines nothing but a feed forward neural network without any tuning and has a single hidden layer. The experimental results on different datasets have shown that the combination of machine learning techniques shows a performance improvement.