Machine learning to explore data - data feature analysis (Pareto analysis)

Pareto analysis, that is, contribution analysis, to find a small number of factors play a crucial or decisive results in all factors.

There is a common law ---- twenty-eight law, businesses are profound meaning in the interpretation of the law:
for example:
1 80% of the company's profits come from 20% of the best-selling products, while the other 80% of the products produced only 20% of the profits;
2 approximately 80% of the world's resources by 20% of the world's population are depleted;
80% 3 20% of the world's wealth is owned by the person;
420% of the population or 20% of the disease , it will consume 80 percent of medical resources.
In some specific areas, twenty-eight law might be derived from the law and even more for the nineteen gap.

In data mining, we also need to find independent variable elements on the dependent variable to the maximum extent based on eight law thoughts.

As seen in the above example, the front 7 profit proportion has reached 80%.
Especially in terms of merchandising, Pareto analysis results can be used to guide businesses to move goods to build on the attention on key products.

Guess you like

Origin www.cnblogs.com/pythonfl/p/12486579.html