Using the Power of Our Artificial Intelligence in Discovering New Products

Recently there has been much talk about Google’s AlphaGo that defeated the best human player in the game of Go. A remarkable step that shows the potential of artificial intelligence. Neil Lawrence from The Guardian had an especially interesting take on the subject as he drew a parallel with the development of the steam engine in 1712. The steam engine was initially built specifically for a single use case, pumping water out of coal mines. However, it was James Watt who made significant improvements in the efficiency of the steam engine’s condenser so that it could be used in many other fields aside from pumping water. Currently, as Lawrence argues, “machine learning is missing its separate condenser moment”.

Over the past months we at CartSkill have been working diligently in our specific domain of using artificial intelligence in image recognition. Plainly speaking it is the way in which computers’ have the ability to see and recognize images. But we have not only focused on developing the technology but have also applied our work into solving specific problems that users have. And it is now exciting to see our work materialize. More specifically, the problem we are helping to solve is for some as difficult as beating a grandmaster at Go?—?picking the perfect handbag.

With retail companies spending the most in the multi-billion digital advertising market across the globe, product discovery in fashion is a large opportunity. And it is likely that any woman you ask is in the process of looking for a fashion garment but has not yet found that “right one”.

As a start we have deployed our product discovery engine using AI that allows the user to find the right handbag. When clicking on a handbag inside a banner ad, the user is recommended three new handbags, which are all similar to the one that she clicked. Clicking again will reveal new recommendations and so on. The neat thing is that while the user is shown only 4 handbags at a time, the recommendations are intelligently made from a selection of 1,200 bags based only on visual similarity. We partnered with a popular lifestyle blogger and 

now you can actually use our artificial intelligence powered product discovery engine in this blog

to discover the perfect handbag with a couple of clicks.

In addition we have been actively working with bloggers and end users. After trying the recommendation engine, Eliise from the fashion product blog Beauty Express told us that she likes it because “it takes into account the type of bags that the user has clicked and recommends based on these selections. It is better than current category based searches in online stores because it saves time by offering relevant results from a wide variety of products”.

Aside from tackling the problem of product discovery, this solution is also tackling the problem of “banner blindness” in digital advertising but more on that in our next posts. Until then, happy discovery.

Note: you can follow our stories on our Medium blog.