Speakers:
It’s Not Just the Ad: How Meta Learned to Rank Everything Around It
Date:
Monday, November 17, 2025
Time:
10:45 am
Track:
Room:
Charlottenburg
Summary:
Most recommendation systems focus on ranking the main item—whether it’s the perfect product, ad, or video. But in high-impact environments like Meta, optimizing the presentation of that item can be just as critical. Which image best complements an ad? Which text overlay (“Best Seller,” “Limited Time”) will drive engagement? These “non-product” recommendations are often overlooked, yet they can be harder to model than the primary recommendation itself.
From Manual to Automated Author Identification: How Springer Nature Finds the Needles in the Haystacks
Date:
Monday, November 17, 2025
Time:
11:15 am
Track:
Summary:
One of the key initiatives at Springer Nature, journal-based Collections, invites authors to submit papers on specific themes. However, identifying relevant people to invite, from millions of authors, was a cumbersome process. To solve this, an ML based algorithm was built, which identifies authors, based on their recent interactions with us. This reduced the processing time from 4–6 hours to a few minutes and fostered precision and scale. Join Bhaswati and Audrey for operational efficiency insights, finding needles in haystacks.