Further to the left, a node will appear. Further to the left, the original horizontal line is split in two. Plus the original three, that is to say, they are divided into three categories, namely Ningxia, Gansu and other regions.
Further to the left, there are four, divided into four categories, Ningxia is the first category, Gansu is the second category, Hainan and Xinjiang are the third category, and others are the fourth category.
Further to the left, it is divided into five categories, namely Ningxia, Gansu, Xinjiang, Hainan and other regions.
In this way, and so on. If you want to divide into several categories, just find a few horizontal lines. From left to right, each sample is classified into a category.