Feel free to skip down the page if you’re just looking for the quick demo on how to do a market basket analysis in Excel.This all makes sense, right? There are 3 ways to measure association:Typically, when you work with the Apriori Algorithm, you define these terms accordingly.Now suppose, after filtering you still have around 5000 items left. Because the associations are sorted on the basis of decreasing lift ratio, these associations are credible as far as confidence or lift ratio is taken into account. Thus, the next level has much fewer associations to be tested. Confidence denotes the precision with which we can say a transaction is true. This rule means that customers who purchase A are likely to purchase B and thus keeping both of them together will result in B selling more. Creating association rules for them is a practically impossible task for anyone. Apriori algorithm assumes that any subset of a frequent itemset must be frequent. You can install the add on for Excel and try out the association rule option. Almost 70% of the customers buy A as one of their products and 80% of the customers buy B as one of their products. This method of cross-selling comes directly from market basket analysis.
One possible approach is demonstrated in the attached spreadsheet. Intuitively, a lift ratio greater than 1 means that if customers buy A, they are more likely to buy B. This is where the apriori algorithm comes into picture.

At the first level, all associations where support and confidences are lower than the set thresholds are eliminated. Overall we see that the confidence for B → A is lower than A → B so if we have to put our bets on one of the two, we will go for A → B association. We also have three association rules where the confidence is 100%. This is what market basket analysis is all about.This is just a small example.

By scanning the database for the first time, we obtain the following result It is based on the concept that a subset of a frequent itemset must also be a frequent itemset.

The confidence for A → B can be calculated using the formula:Using Support(A and B) = 60% and support(A) as 70%, we get the confidence as 6/7 which is 85.7%.
Product placements should be done in such a way that the items frequently bought together are kept next to each other, so that customers are encouraged to buy them and so that this results in a boost in sales.One very famous example of market basket analysis has to do with beer purchases and diapers. Hi Does anyone know if it is possible to use VBA to write an 'Apriori Algorithim' and if so any idea how? Apriori Algorithm is an exhaustive algorithm, so it gives satisfactory results to mine all the rules within specified confidence and sport. Suppose you made a rule about an item, you still have around 9999 items to consider for rule-making. Since there are 50 items for 75,000 members, I will use 1,500 transactions as the minimum support(2%). we can also illustrate this through a variety of examples. Do we have the association of A → B or B → A?We know that all the customers do not buy A. In Excel kann über die statistischen Analyse-Funktionen eine Korrelationsmatrix erstellt werden, wie im Teil 1 der Blogseriegezeigt. So before we understand the Apriori Algorithm, let’s understand the math behind it. For retailers, understanding this kind of customer behavior can result in a boost in sales. This means that the lift ratio is the ratio of support of all items occurring together to the support of each of them independently. Output ist eine Tabelle, in der die Korrelationen aller Produkte mit allen anderen berechnet werden.