This is a method of making automatic predictions (filtering) about the interests of a shopper by collecting preferences or taste information from many users (collaborating). The underlying assumption of the collaborative filtering approach is that if a person A has the same opinion as a person B on an issue, A is more likely to have B’s opinion on a different issue than that of a randomly chosen person. In the more general sense, collaborative filtering is the process of filtering for information or patterns using techniques involving collaboration among multiple agents, viewpoints, data sources, etc.
There are two types of collaborative filtering:
Collaborative Filtering [CF] : This involves finding the best products to display within a carousel, on a given product page, to preserve the interest of a customer during their journey in an eCommerce website & enhance relevant product discovery. The algorithm finds the typical prospective products that a customer might showcase interest in, based on a source/base product within the same category.
Cross-Product Collaborative Filtering [CP] : Same as the above, the only difference being, that the recommendations are done for other categories that are affine to the source/base category. These recommendation can be shown in a product page, in the cart page or at the point when an order has been placed. This is analogous to cross-selling in retail, meant to enhance cross category product discovery & enhanced customer experience.