What started in 2012 as a way to authenticate art has now progressed into the only technology-based, powered by AI, authentication solution on the market to provide trust to buyers and sellers of luxury handbags. Entrupy has collected thousands of images of physical goods (bags, shoes, watches, jewelry, etc) from around the world, creating a rich and diverse unmatched data-set.
We utilize this high quality data collected to teach our algorithms to differentiate between fake and authentic items, which in turn, gives us the ability to produce results like our 99.1% accuracy rate. Every item that’s authenticated helps the algorithms learn and improve, creating a smarter solution that adapts to the changing world.
Where and how is the original data collected?
Our proprietary data has been collected over years and from diverse sources from around the world. We also have data partnerships with prominent businesses to collect data on their inventory, thus assuring us pre-checked items.
How do you know that your original dataset is correct?
We run 3 different layers (2 manual verifications and one algorithmic) of filtering that allows us to correct any possible errors in the data submitted. This ensures that the algorithms are always fed extremely high quality information during the machine-training process.
Does this cover only recently available items and how do you keep updated with the new releases?
In general, we cover most styles of a particular brand or type of material. We have data samples from over 80 years ago all the way up to the recent months. This allows us to cover a wide time-range for most supported brands. In order to provide our customers with seamless authentication ability, we keep track of the recent releases and try to incorporate them as soon as possible.
To learn more about our technology, the thought process behind the building of our technology and a sneak peek into our capabilities, below is an excerpt of the paper published by the founders of Entrupy.