Provide a unique, personalized experience to every shopper
With a few questions, we determine the user’s shopping preferences as well as their right size and fit,
creating a customized shopping experience where the user only sees relevant items.
We use a multi agent recommendation engine based on individual and collective user interactions, user
measurements, user clustering and user and product similarity, that learns over time.
For user clustering and similarity we developed an algorithm based on:
— A Vectorial space where we manage all our users and products.
— A K-medioids algorithm for
clustering users and products.
— A K nearest neighbours algorithm for finding similar users.
For finding product similarity we use:
— A Neural network for finding 1550 key components in images.
— A Multidimensional vectorial space
for measuring similarity between garments in an euclidean space.