Clustering by growing incremental self-organizing neural network

作者：

Highlights：

• A topological graph visualizes both data partitions and details of each cluster.

• The presented approach aims to learn data distribution of each cluster.

• The number of clusters is not a prerequisite for clustering tasks.

• The proposed method is able to detect arbitrary-shaped clusters.

摘要

•A topological graph visualizes both data partitions and details of each cluster.•The presented approach aims to learn data distribution of each cluster.•The number of clusters is not a prerequisite for clustering tasks.•The proposed method is able to detect arbitrary-shaped clusters.

论文关键词：Clustering,Unsupervised learning,Self-organizing neural networks,Incremental learning,Data visualization

论文评审过程：Available online 16 February 2015.

论文官网地址：https://doi.org/10.1016/j.eswa.2015.02.006