Combating User Fatigue with Frequency-Recency Features
Senior Data Scientist, TL of Data Science – Outbrain
Combating User Fatigue with Frequency-Recency Features
Senior Data Scientist, TL of Data Science – Outbrain
Natalia is a senior data scientist and data science team leader at Outbrain in the Recommendations group. She is involved in the design and improvements of algorithms that control personalized ad selection.
Before joining Outbrain, Natalia was a research scientist at Yahoo Research Haifa in the Native Ad Science group. Before joining that group, she was with the Mail Mining group, primarily analyzing and modeling mail data to devise novel mail features.
Natalia completed her PhD on “Coding Theory and Projective Spaces” at the Technion, Computer Science Department, under the supervision of Prof. Tuvi Etzion in 2011. She then joined Prof. Sriram Vishwanath’s Wireless Networking and Communications Group at the Department of Electrical & Computer Engineering, University of Texas, Austin, as a postdoctoral fellow, focusing on coding for distributed storage systems
Natalia is a senior data scientist and data science team leader at Outbrain in the Recommendations group. She is involved in the design and improvements of algorithms that control personalized ad selection.
Before joining Outbrain, Natalia was a research scientist at Yahoo Research Haifa in the Native Ad Science group. Before joining that group, she was with the Mail Mining group, primarily analyzing and modeling mail data to devise novel mail features.
Natalia completed her PhD on “Coding Theory and Projective Spaces” at the Technion, Computer Science Department, under the supervision of Prof. Tuvi Etzion in 2011. She then joined Prof. Sriram Vishwanath’s Wireless Networking and Communications Group at the Department of Electrical & Computer Engineering, University of Texas, Austin, as a postdoctoral fellow, focusing on coding for distributed storage systems
Outbrain serves millions of users with personalized native ads and recommendations. One of the most challenging issues in such recommender systems is finding the optimal balance between user satisfaction and overall revenue. When users see the same ad repetitively, it hurts user experience the most. This problem leads to the phenomenon called “user fatigue,” expressed in a decrease of the click-through rate with every impression of the same ad. However, naive solutions to this problem may come at the cost of reduced revenue.
An elegant solution incorporates new features based on recent historical interactions of users with ads into the click prediction model. Specifically, the approach combines information about the frequency and recency of past user-ad interactions in a sense that very recent interactions (that happen in recently) get higher importance than those that happen a few hours or days prior.
The features were implemented and tested thoroughly in online A/B testing systems and showed that this approach reduces user fatigue by diversification of displayed ads and increases revenue.
Outbrain serves millions of users with personalized native ads and recommendations. One of the most challenging issues in such recommender systems is finding the optimal balance between user satisfaction and overall revenue. When users see the same ad repetitively, it hurts user experience the most. This problem leads to the phenomenon called “user fatigue,” expressed in a decrease of the click-through rate with every impression of the same ad. However, naive solutions to this problem may come at the cost of reduced revenue.
An elegant solution incorporates new features based on recent historical interactions of users with ads into the click prediction model. Specifically, the approach combines information about the frequency and recency of past user-ad interactions in a sense that very recent interactions (that happen in recently) get higher importance than those that happen a few hours or days prior.
The features were implemented and tested thoroughly in online A/B testing systems and showed that this approach reduces user fatigue by diversification of displayed ads and increases revenue.
8:45 | Reception |
---|---|
9:30 | Opening words by WiDS TLV ambassador Nitzan Gado and by Lily Ben Ami, CEO of the Michal Sela Forum |
9:50 | Prof. Bracha Shapira – Data Challenges in Recommender Systems Research: Insights from Bundle Recommendation |
10:20 | Juan Liu – Accounting Automation: Making Accounting Easier So That People Can Forget About It |
10:50 | Break |
11:00 | Lightning talks |
12:20 | Lunch & poster session |
---|---|
13:20 | Roundtable session & poster session |
14:05 | Roundtable closure |
14:20 | Break |
14:30 | Merav Mofaz – “Every Breath You Take and Every Move You Make…I'll Be Watching You:” The Sensitive Side of Smartwatches |
14:50 | Reut Yaniv – Ad Serving in the Online Geo Space Along Routes |
15:10 | Rachel Wities - It’s Not Just the Doctor’s Handwriting: Challenges and Opportunities in Healthcare NLP |
15:30 | Closing remarks |
15:40 | End |
WiDS Tel Aviv is an independent event that is organized by Intuit’s WiDS TLV ambassadors as part of the annual WiDS Worldwide conference, the WiDS Datathon, and an estimated 200 WiDS Regional Events worldwide. Everyone is invited to attend all WiDS conference and WiDS Datathon Workshop events which feature outstanding women doing outstanding work.
© 2018-2023 WiDS TLV – Intuit. All rights reserved.
Scotty – By Nir Azoulay
Design: Sharon Geva