Home > AI Solutions > Artificial Intelligence > White Papers > Using Machine Learning to Recommend Service Product Offerings > MBA-based service recommendation
MBA attempts to identify item sets that are frequently purchased together. Because Dell stores a tremendous quantity of order data, we can use MBA to derive association rules for asset and service offer combinations (Figure 5). For example, if a customer has purchased product line A, we can recommend that they may also want to buy a Tier 2 warranty. We are 68.18% confident that this recommendation will be accepted, based on 972 previous orders in our dataset. “Lift” is a measure of the association strength between items. If the lift value is less than or equal to 1, there is no association between the item sets. If the lift value is greater than 1, there is a strong association between item sets. In this example, the lift value is 1.56, indicating an association between product line A and the Tier 2 warranty.