What else did people who bought a certain item buy? What are the key points of realizing this function?

1. For example, a bought item A and item B today, he placed this order, and then when b bought item A, item B was recommended below.
Or maybe a bought item A today and placed an order, and bought item B the next day. When b bought item A, item B was recommended below.

Is this idea reasonable? Is there anything else that needs to be thought about in depth about this feature?

Mobile editing (+_+)

Reply content:

1. For example, a bought item A and item B today, he placed this order, and then when b bought item A, item B was recommended below.
Or maybe a bought item A today and placed an order, and bought item B the next day. When b bought item A, item B was recommended below.

Is this idea reasonable? Is there anything else that needs to be thought about in depth about this feature?

Mobile editing (+_+)

Analyze a large number of user behaviors and then give suggestions. This is a big data application. The key to big data applications lies in statistical data, not the data of a certain individual.

As for this function, you can count the people who bought A (recently) and what they bought in the nearby period, add them up by person (or person-time), and take out the top ones.

Of course, this simple algorithm may not be able to meet your actual needs. What you actually need needs to be analyzed by yourself and add some conditions, such as adding item classification, promotion activities and other factors for analysis