NG2CE: Double neural gas based cluster ensemble framework

Though there exist different kinds of cluster ensemble methods, few of them consider how to process the dataset with noisy attributes. In this paper, we design a double neural gas based cluster ensemble framework, named as NG
²
CE, which is a new cluster ensemble framework to process the dataset with noisy attributes. Compared with traditional cluster ensemble methods, the cluster framework NG
²
CE not only adopts the neural gas algorithm to perform clustering on the data dimension, but also applies the neural gas algorithm to the attribute dimension, which will increase the diversity of the ensemble and improve the accuracy of the final result. NG
²
CE also adopts the normalized cut algorithm to partition the consensus matrix and obtains the final result. The results of the experiments on both synthetic datasets and real datasets show that (i) NG
²
CE works well on both synthetic datasets and real datasets. (ii) NG
²
CE is able to improve the accuracy and the stableness of the final result.