A great example from the online grocery store is how the Raptor Smart Advisor recognizes patterns in user behavior. When consumers buy hypoallergenic laundry detergent, the algorithm automatically identifies various preferences through the purchase.
In the example above, our algorithm recommends products that other consumers have bought with that exact laundry detergent. In this case, the related products are other laundry products and cleaning supplies, all hypoallergenic.
Based on patterns in user behavior, we can see that consumers, who buy hypoallergenic detergent, also have preferences for other cleaning products to be allergy friendly.
Logically as it may sound, this automated process saves nemlig.com several hours of manual tagging.
A second example is how the algorithm detects patterns that can seem surprising to most people. For example, you might not realize the connection between “Boil-In-Bag” rice, pre-cooked meals and oatmeal, but the data shows a clear relation in purchase history.
When considered, all those products show an affinity for quick and easy meal preparation. However, if you click on Risotto rice, you will see ingredients that reflect a preference for meticulous cooking.
The recommendations will always reflect the behavior of the consumers, which provides the most relevant recommendations while simultaneously freeing up time at nemlig.com.