Improving k-Means Clustering with Disentanglement

Learning the dataset class neighborhood structure improves clustering

An accompanying article for the paper “Improving k-Means Clustering with Disentangled Internal Representations” by A.F. Agarap and A.P. Azcarraga presented at the 2020 International Joint Conference on Neural Networks (IJCNN)

Background

Clustering is an unsupervised learning task that groups a set of objects in a way that the objects in a group share more similarities among them than those from other groups. It is a widely-studied task as its applications …

article artificial intelligence clustering conference data science dataset deep learning international k-means machine learning networks neural networks objects paper set thoughts-and-theory unsupervised unsupervised learning