What type of study is this?

This is a Experimental Study study.

A Reward-Informed Semi-Personalized Bandit Approach for Enhancing Accuracy and Serendipity in Online Slate Recommendations

View Full Paper

Ask AI

Bookmark

Authors

Observational analysis improved serendipity in online recommendations, suggesting semi-personalized bandits mitigate overspecialization.

A semi-personalized bandit approach reduces average regret in recommendation systems, promoting diverse content exposure.
Experiments demonstrated that personalization can lead to overspecialization, but integrating serendipity yields better recommendations.
Using decision trees for user segmentation allows for improved recommendation accuracy and unexpected suggestions.
The framework relies on Thompson Sampling for effective exploration, reinforcing its applicability in complex online environments.

Kerpel et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark