Читать книгу Data Analytics in Bioinformatics - Группа авторов - Страница 45

This follows divisive clustering approach in which the whole dataset is divided into 2 sub parts (clusters) and these clusters are further separated into sub clusters until no splitting can be performed further [17].

Drawback—once separated clusters cannot be merged, this cannot handle huge dataset [19]. This may not be applicable for gene expression data as it cannot handle missing data [17].