An Ensemble Clustering Model For Dimension Reduction

Throughout the data explosion era, dimension reduction is a vital area of machine learning techniques to achieve useful and reduced data-sets. on the other side, ensemble models have consensus mechanism to take advantage of positive point of several clustering techniques concurrently between the various clustering algorithms that suffer from negative aspects. in this study, we use an ensemble clustering model with k-means to aim dimension reduction, two co-association matrix, two consensus functions to aggregate clustering results, also pca. the model significantly reduces data-sets dimensions by feature extraction techniques. we applied nmi performance validity index to evaluate results. the simulation results show this model acquires better clustering performance for all data-sets while accomplish feature extraction.