collaborative filtering (CF)

Collaborative filtering(CF) uses the preferences of an entire group to draw conclusions about the interests of an individual. It is a mathematical approach that uses a lot of information about the behavioral pattern of a group of buyers, a collaboration, to specifically examine the buying behavior of an individual without explicitly knowing his behavior.

Collaborative filtering uses correlation to identify similarities between users. CF data analysis is a form of data mining in which specific data is filtered from a pool of information to infer user interests. The filtering technique is used to personalize an individual's buying behavior. For this purpose, the purchasing behavior of customer groups is analyzed, their behavioral patterns are examined and these are placed in relation to the personal data. The aim is to create a personal profile that can be used for sales activities such as online advertising, making them more efficient. In extended CF concepts, even predictions about user interest and their preferred decision patterns should be made.

For example, collaborative filtering can be used to determine gender, age, and income structure, among other factors, and this information can be used to display an appropriate web advertisement. For example, a smaller or larger car in a color preferred by the target group. A well-known CF example can be found at Amazon, which derives conclusions about other book topics from the purchase of a book title and recommends them to the buyer.