Basket analysis - a method of data mining, which creates a set of approximate associative rules for the data set of the type "If the predecessor is usually the successor", where the predecessor and the consequent are logical expressions, eg If (car_type = 'sports' and age & lt; 25) then usually (risk = 'high' and insurance = 'high')
Basket analysis is particularly well suited for processing nominal and order variables, so it is often used in text mining where nominal variables are dominant.
Another common use of basket analysis (from which it has its name) is to examine customer p (that is, what they put into the shopping cart). Analyzing the basket allows you to get rules like "If a customer buys a diaper and buys a toy, it will often buy blue clothes", which allows for example to optimize the layout of the merchandise in the shop.
An efficient and popular algorithm for basket analysis is the apriori algorithm (Agrawal and Swami, 1993; Agrawal and Srikant, 1994; Han and Lakshmanan, 2001; Witten and Frank, 2000). Bibliography
wiki
Comments
Post a Comment