Market Basket Analysis

getwd()

## [1] "C:/Users/user/Documents"

Collaborative Filting

Collaborative Filting is the most commonly used recommendation system. The most typical usage of it is shown in this figure. Basically, based on the previous purchase of a person, related items are recommended. However, the assumption of this approach that we know very well, via enough history, to predict this customer’s behavor.

Market Basket Recommendation

Market basket recommendation

Figure

In the market basket analysis, however, a totally different approach is taken. It is not based on item relationships, rather, is based on the association. The central part in building a recommendation engine is the Association Rule Learner node, which implements the Apriori algorithm, illustrated below. In the traditional retail business, the two nodes are: inventory and point-of-sale data, to predict which combinations of products will sell the best.

https://stats.stackexchange.com/questions/256012/item-item-collaborative-filtering-vs-market-basket-analysis https://www.knime.com/knime-applications/market-basket-analysis-and-recommendation-engines https://nealanalytics.com/understanding-market-basket-analysis-product-recommendation/