How is E-Commerce Using AI to Make Recommendations?

David MillerGeneral

The nuts and bolts of AI in e-commerce lie in Convolutional Neural Networks

AI is a generally used term for what is actually a build up of several components and algorithms. The convolutional neural network makes the applications of recommendations in e-commerce feasible by learning to recognize image representation and find similar images. Based on this ability to learn images and search for similar products, the CNN can make applicable recommendations. It is an algorithm that was inspired by the complexities of the animal visual cortex. It is comprised of several modules whose sole purpose are to identify and analyze visual imagery.

If the network is designed to work within facial recognition, then some modules may be activated when they spot various facial features (ear, nose, etc.). Convolutional Neural Networks are viewed as being among the most modern advances when working with pixels. To get a better idea, go to Google and type “blue jeans” into the search field and click “images”. From the results that you see you should have a pretty good idea of what CNN is capable of doing.

How this helps e-commerce give the ultimate customer experience

Blue jeans will continue to be the subject matter to better help create an understanding of what the customer will experience while shopping or simply browsing online fashion.

So, you are browsing your favorite online fashion retailer as you are in need of a new pair of blue jeans that you can wear for casual engagements as well as informal business events. You see a nice pair of blue jeans but they are just not the exact item that you need. Well, now imagine if someone was able to offer you blue jeans that are similar but had some small differences that may better suit you. With AI, more specifically CNN, it is absolutely possible.

Convolutional Neural Networks

Convolutional Neural Network identified similar blue jeans


One could ask if it would be better to just take a random pair of jeans from the same category. To elaborate further, say that you also desire having the same color or tint of the original jeans viewed. Take a look at how the algorithm would perform in this case.

Recommendations Exponea

similar products based on simple rules

Similar products based on simple rules

You can see that the first row is comprised of random images from the blue jeans category. Meanwhile, the second row was restricted to the color blue. This illustrates that the imperfections in data labelling, based on rules, can disrupt algorithms. No algorithm is without faults but those built on rule based recommendations are not as strong as those designed around CNN based AI.

E-commerce companies grow because of AI

Quite simply put, e-commerce companies who use AI have a strong competitive edge over those who do not. Not many software companies are able to offer the online fashion industry the AI that they actually need, particularly with the inclusion of Convolutional Neural Networks.

AI as a whole has useful applications that customers are seeing as an advantage. That is why they are spending more money online, year over year.

E-commerce companies that have integrated with Exponea as their AI-driven marketing platform allow for several things to happen – including happier customers, higher conversion rates, and all around increased revenue. Learn more about how Exponea Experience Cloud recommendations are able to scale your online business while creating happier customers.

How artificial intelligence  revolutionizes online fashion retail