With the data explosion more and more data are collected from multiple sources represented by multiple views, where each describes a perspective of the data. To deal with this kind of data in the context of unsupervised learning, one can rely on factorial approaches and clustering. Depending on the objective, these two types of methods can be used separately, successively in a two step approach or simultaneously leading to subspace clustering. In this presentation, we will review, discuss and illustrate different unsupervised approaches from the most classical to the most recent.