W. Maréchal. Medium, September 24th 2022.
Abstract: Clustering is a reduction of available data into a manageable number of choices, either by similarity or by defining a distance between the data points. It is a vastly used machine learning method, and the more it is used, the more branches and leaves it has. But the documentation often contains just the mathematical expression, leaving out the meaning, and several names can be associated with the same expression. Not knowing which one to use, the default one, or the one used by our peers is used. The scikit-learn organization proposes on its website a comparison of the main methods on several data sets and provides an explanation for the workings of the algorithms.
Comments