Using Images in Recommender Systems
summary
The talk surveys how images are used inside recommender systems, covering techniques for incorporating visual information to influence click decisions and improve recommendations. It also discusses methods to reduce computational load when processing images in high‑traffic production systems.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:08Simple approaches without images
- 2:55Game data availability
- 3:38Image influence hypotheses
- 4:21Experiment design
- 5:02Beauty image hypothesis
- 6:48Text on images hypothesis
- 7:11Neural network approaches
- 8:56Model selection for recommendations
- 11:26Metrics and validation
- 12:52Light FM results
- 13:56Resource optimization
- 16:09Further improvements
- 17:15Conclusion and Q&A
related talks
41mRecommender Systems: Crafting Effective Personalised Experiences
34mRecommender Systems
58mRecommender Systems – A Walkthrough
50mRecommender Systems: What They Are and Where to Find Them
32mRecommender Engines - What are they? Where do they live? What do they feed on?
49mAn Introduction to Recommendation Engines
17mCalculating recommendations based on product images #bbuzz
21mUnleashing the Power of Data Science