|Mon Aug 22 2016 22:03:51|
eBay Unleashes Machine Learning on Search Pages
By: Ina Steiner
Everyone is buzzing about Artificial Intelligence and machine learning these days, and eBay is no exception. Now comes the revelation that eBay data scientists are using the technology in its Best Match search algorithm.
In a post on Monday, eBay wrote, "The largest scale application of machine learning technology at eBay is currently Best Match, the algorithm used to optimize relevance for buyers during their shopping experiences."
Best Match is a sore point for many sellers. They want eBay to return listings in search results solely on relevance, but the marketplace uses many other factors. For instance, eBay uses Best Match to reward sellers for whatever behavior it wants sellers to follow, from free shipping to generous return policies.
In 2014, we published a column
written by a seller who believed he had cracked the code on eBay search - it gives a good idea as to some of the factors eBay may be using when ranking listings on its search results pages.
But with data scientists let loose on Best Match, things are more complicated than ever. For instance, in Monday's post, eBay data scientist Alex Cozzi said, "We are also applying statistical learning to optimize the whole page: reacting to the users' actions we can reorder and prioritize the content in the search page, providing search guidance or access to top deals and top products in appropriate context while minimizing distractions."
As we point out in Tuesday's Newsflash
, that means the results a seller sees may be very different from what other buyers and sellers see - and that can make search optimization quite challenging.