An Improved Document Clustering Algorithm Based on Neural Gas Algorithm

In this paper, we put forward a new document clustering algorithm by improving the existing Neural Gas algorithm. Our insight is that the degree of effect of any point on a cluster centroid depends on the distance values between this point and the other more near cluster centroids. Experimental results on a number of real-life data sets suggest that our algorithm outperforms other clustering algorithms in terms of five indices: Entropy, Purity, F1 values, Rand Index, and normalized mutual information (NMI ).